{"run":{"job_id":"86271","host":"GPU-830","model_id":"/mnt/ceph3/ec/jacky/sam3.1","propose_model":"gemini-3.5-flash-lite","select_model":"gemini-3.5-flash-lite","sample_fps":4.0,"choice_frames":8,"clips_requested":20,"output_dir":"/home/zzk/data/video_bbox/vlm_gemini3.5_lite","wall_seconds":553.82,"measured":{"mean_clip_duration_seconds":17.8658425,"frames_per_clip":69.4,"tracks_per_clip":3.75,"noun_phrase_gate_survival":1.0,"verification_yield":0.717391304347826},"rejections":{"no_candidates_retrieved":6,"count_mismatch":5,"implausible_box":2}},"source":{"bucket":"mv-ytun-external1","prefix":"silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/","category":"test_vehicles","filters":"extension .mp4; storage class not GLACIER or DEEP_ARCHIVE","shuffle_seed":0,"clips":["s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_003_scene_023.mp4","s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_000_scene_044.mp4","s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_014_scene_002.mp4","s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_008_scene_014.mp4","s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_001_scene_010.mp4","s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_006_scene_028.mp4","s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_014_scene_013.mp4","s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--Dx9HVcQWs_chunk_002_scene_013.mp4","s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--Dx9HVcQWs_chunk_001_scene_034.mp4","s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_015_scene_004.mp4","s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_012_scene_010.mp4","s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_012_scene_017.mp4","s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_000_scene_042.mp4","s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_011_scene_019.mp4","s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_011_scene_001.mp4","s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_010_scene_016.mp4","s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--X6fL8DOMM_chunk_000_scene_003.mp4","s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_011_scene_010.mp4","s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_012_scene_020.mp4","s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_014_scene_018.mp4"]},"clips":[{"index":1,"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_003_scene_023.mp4","status":"complete","steps":{"decode":{"inputs":{"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_003_scene_023.mp4","sample_fps":4.0},"outputs":{"width":1920,"height":1080,"source_fps":50.0,"duration_seconds":32.0,"frames_kept":124,"blank_frames_dropped":0,"timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88,10.14,10.4,10.66,10.92,11.18,11.44,11.700000000000001,11.96,12.22,12.48,12.74,13.0,13.26,13.52,13.780000000000001,14.040000000000001,14.3,14.56,14.82,15.08,15.34,15.6,15.860000000000001,16.12,16.38,16.64,16.9,17.16,17.42,17.68,17.94,18.2,18.46,18.72,18.98,19.240000000000002,19.5,19.76,20.02,20.28,20.54,20.8,21.06,21.32,21.580000000000002,21.84,22.1,22.36,22.62,22.88,23.14,23.400000000000002,23.66,23.92,24.18,24.44,24.7,24.96,25.22,25.48,25.740000000000002,26.0,26.26,26.52,26.78,27.04,27.3,27.560000000000002,27.82,28.080000000000002,28.34,28.6,28.86,29.12,29.38,29.64,29.900000000000002,30.16,30.42,30.68,30.94,31.2,31.46,31.720000000000002,31.98],"error":""}},"propose":{"inputs":{"model":"gemini-3.5-flash-lite","temperature":0.4,"video_bytes":54298085,"system_prompt":"You are annotating video clips to build training data for referring multi-object tracking: given a video and a natural-language description, a model must output the boxes of every object matching that description, over time.\n\nYour job is to write the descriptions. You do NOT output coordinates, boxes, pixel positions, or any numbers about where things are. A separate detector finds the objects.\n\nFor every clip you produce a list of items. Each item has two texts that must NOT be the same:\n\nEXPRESSION — what a person would say to single out the object or objects. It may describe motion, temporal order, interaction with other objects, and spatial relations. It should read like something a person would actually say.\n  Good: \"the black car that's turning left\"\n  Good: \"the cyclist who overtakes the bus\"\n  Good: \"the two people who walk out of frame on the right\"\n\nNOUN_PHRASE — the same object as a plain, timeless visual category with its visible attributes, and NOTHING else. The detector is prompted with this text and looks at ONE frame at a time, so anything about time, order, or action is useless to it and makes it fail.\n  Allowed: category nouns, colours, materials, clothing, size, type\n  Forbidden: verbs and participles of action (\"turning\", \"walks\", \"is moving\"), temporal words (\"then\", \"starts\", \"after\", \"first\", \"last\"), relative clauses (\"that ...\", \"which ...\", \"who ...\"), ordinals and rank words (\"first\", \"leftmost\", \"nearest\"), and anything over six words.\n  \"the black car that's turning left\" -> \"black car\"\n  \"the cyclist who overtakes the bus\" -> \"cyclist\"\n  \"the two people who walk out of frame on the right\" -> \"person\"\n\nKIND — what makes the expression hard: \"motion\", \"appearance\", \"spatial\", or \"interaction\". Aim for about 35% \"motion\" items: motion is the part a single-frame model cannot already solve, so it is the most valuable, but a list of only motion items is not usable training data either.\n\nEXPECTED_COUNT — how many distinct objects in the clip match the EXPRESSION. Count carefully; this is checked against the detector's answer, and a wrong count discards the item.\n\nIS_NEGATIVE — at most ONE item may be a negative, and only when you also write at least one non-negative item. A negative is an expression that recombines attributes that ARE present in the clip into a whole that is NOT. If the only black car is moving, \"the parked black car\" is a good negative. Its NOUN_PHRASE must still be the category that IS present (\"black car\"), because that is what makes the absence checkable rather than merely unverified. A negative's EXPECTED_COUNT is 0.\n\nRules that discard an item, so follow them exactly:\n- Never write coordinates, box values, pixel positions, or percentages of the frame.\n- Never let a NOUN_PHRASE contain a verb of action, a temporal word, a relative clause, or an ordinal.\n- Never write an EXPRESSION whose object is not visible in the clip, unless it is marked IS_NEGATIVE.\n- Prefer objects that are clearly visible and identifiable over background clutter.\n- Do not describe the camera or the shot; describe objects in the scene.\n","user_prompt":"Watch this clip and write between 1 and 3 items following the rules. Cover different objects rather than restating one object several ways, and include any negative in the same list.\n"},"outputs":{"raw_response":"{\n  \"items\": [\n    {\n      \"expression\": \"the black and white dog walking on the stone pavement\",\n      \"noun_phrase\": \"black and white dog\",\n      \"kind\": \"motion\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the brown dog walking up the stone stairs in the background\",\n      \"noun_phrase\": \"brown dog\",\n      \"kind\": \"motion\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    }\n  ]\n}","raw_items":[{"expression":"the black and white dog walking on the stone pavement","noun_phrase":"black and white dog","kind":"motion","expected_count":1,"is_negative":false},{"expression":"the brown dog walking up the stone stairs in the background","noun_phrase":"brown dog","kind":"motion","expected_count":1,"is_negative":false}],"accepted":[{"expression":"the black and white dog walking on the stone pavement","noun_phrase":"black and white dog","kind":"motion","expected_count":1,"is_negative":false},{"expression":"the brown dog walking up the stone stairs in the background","noun_phrase":"brown dog","kind":"motion","expected_count":1,"is_negative":false}],"gated":[]}},"track":{"inputs":{"tracker":"Sam31MultiplexTracker","model_id":"/mnt/ceph3/ec/jacky/sam3.1","noun_phrases":["black and white dog","brown dog"],"expressions_in":2,"frames":124,"sample_fps":4.0,"frame_size":"1920x1080","timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88,10.14,10.4,10.66,10.92,11.18,11.44,11.700000000000001,11.96,12.22,12.48,12.74,13.0,13.26,13.52,13.780000000000001,14.040000000000001,14.3,14.56,14.82,15.08,15.34,15.6,15.860000000000001,16.12,16.38,16.64,16.9,17.16,17.42,17.68,17.94,18.2,18.46,18.72,18.98,19.240000000000002,19.5,19.76,20.02,20.28,20.54,20.8,21.06,21.32,21.580000000000002,21.84,22.1,22.36,22.62,22.88,23.14,23.400000000000002,23.66,23.92,24.18,24.44,24.7,24.96,25.22,25.48,25.740000000000002,26.0,26.26,26.52,26.78,27.04,27.3,27.560000000000002,27.82,28.080000000000002,28.34,28.6,28.86,29.12,29.38,29.64,29.900000000000002,30.16,30.42,30.68,30.94,31.2,31.46,31.720000000000002,31.98],"session":"one session over 124 JPEG frames; then per phrase: reset_session, add_prompt(text) on frame 0, propagate_in_video","passes":2},"outputs":{"raw_tracks":[{"noun_phrase":"black and white dog","object_id":0,"observations":[{"timestamp":0.0,"box":[0.37291669845581055,0.4851851761341095,0.6661458909511566,0.8083333075046539],"occluded":false},{"timestamp":0.26,"box":[0.33385419845581055,0.470370352268219,0.6234375536441803,0.8083333075046539],"occluded":false},{"timestamp":0.52,"box":[0.2953125238418579,0.45185184478759766,0.5906250476837158,0.8620370328426361],"occluded":false},{"timestamp":0.78,"box":[0.2526041865348816,0.43981480598449707,0.572916716337204,0.852777749300003],"occluded":false},{"timestamp":1.04,"box":[0.21562501788139343,0.45370370149612427,0.5661458671092987,0.8620370328426361],"occluded":false},{"timestamp":1.3,"box":[0.19479167461395264,0.44722220301628113,0.5682291984558105,0.8481481075286865],"occluded":false},{"timestamp":1.56,"box":[0.1848958432674408,0.43703702092170715,0.5489583611488342,0.8175925612449646],"occluded":false},{"timestamp":1.82,"box":[0.16718751192092896,0.4120370149612427,0.5359375178813934,0.8324073851108551],"occluded":false},{"timestamp":2.08,"box":[0.13593751192092896,0.4259259104728699,0.5302083790302277,0.8175925612449646],"occluded":false},{"timestamp":2.34,"box":[0.12031250447034836,0.4157407283782959,0.5250000283122063,0.8564814627170563],"occluded":false},{"timestamp":2.6,"box":[0.1640625149011612,0.40833333134651184,0.5609375387430191,0.8148148059844971],"occluded":false},{"timestamp":2.86,"box":[0.24947917461395264,0.3305555582046509,0.6994791924953461,0.9990740418434143],"occluded":false},{"timestamp":3.12,"box":[0.3281250298023224,0.3472222089767456,0.744791716337204,0.7435185015201569],"occluded":false},{"timestamp":3.38,"box":[0.3921875059604645,0.43611109256744385,0.7968750298023224,0.8120370209217072],"occluded":false},{"timestamp":3.64,"box":[0.40677085518836975,0.4861111044883728,0.7708333730697632,0.8527777791023254],"occluded":false},{"timestamp":3.9,"box":[0.39531251788139343,0.5175926089286804,0.7067708671092987,0.8666666746139526],"occluded":false},{"timestamp":4.16,"box":[0.39635419845581055,0.5046296119689941,0.6640625596046448,0.8425925672054291],"occluded":false},{"timestamp":4.42,"box":[0.4208333492279053,0.4888888895511627,0.6505208611488342,0.8638888895511627],"occluded":false},{"timestamp":4.68,"box":[0.43541669845581055,0.47592592239379883,0.6348958760499954,0.8759259283542633],"occluded":false},{"timestamp":4.94,"box":[0.4166666865348816,0.46759259700775146,0.602083370089531,0.8935185074806213],"occluded":false},{"timestamp":5.2,"box":[0.40000003576278687,0.44814813137054443,0.5354167073965073,0.912962943315506],"occluded":false},{"timestamp":5.46,"box":[0.3348958492279053,0.5046296119689941,0.49114586412906647,0.8666666448116302],"occluded":false},{"timestamp":15.6,"box":[0.46562501788139343,0.375,0.5203125216066837,0.43518518283963203],"occluded":false}],"scores":[0.9452055096626282,0.9452055096626282,0.9452055096626282,0.9452055096626282,0.9452055096626282,0.9452055096626282,0.9452055096626282,0.9452055096626282,0.9452055096626282,0.9452055096626282,0.9452055096626282,0.9452055096626282,0.9452055096626282,0.9452055096626282,0.9452055096626282,0.9452055096626282,0.9452055096626282,0.9452055096626282,0.9452055096626282,0.9452055096626282,0.9452055096626282,0.9452055096626282,0.9452055096626282]},{"noun_phrase":"brown dog","object_id":0,"observations":[{"timestamp":0.26,"box":[0.20416668057441711,0.11944444477558136,0.24895834922790527,0.17777777835726738],"occluded":false},{"timestamp":0.52,"box":[0.23541668057441711,0.11203703284263611,0.28072918206453323,0.1657407358288765],"occluded":false},{"timestamp":0.78,"box":[0.2838541865348816,0.10370370000600815,0.32083335518836975,0.1509259231388569],"occluded":false},{"timestamp":1.04,"box":[0.3296875059604645,0.07222221791744232,0.35989584028720856,0.12129629030823708],"occluded":false},{"timestamp":1.3,"box":[0.3812500238418579,0.04814814776182175,0.40312502533197403,0.0981481485068798],"occluded":false},{"timestamp":1.56,"box":[0.43906253576278687,0.032407406717538834,0.4598958697170019,0.06851851567626],"occluded":false},{"timestamp":5.98,"box":[0.7026041746139526,0.031481482088565826,0.7723958417773247,0.06481481343507767],"occluded":false},{"timestamp":6.24,"box":[0.6427083611488342,0.14444443583488464,0.7026041969656944,0.19074073061347008],"occluded":false},{"timestamp":6.5,"box":[0.5880208611488342,0.2611111104488373,0.6463541984558105,0.316666666418314],"occluded":false},{"timestamp":6.76,"box":[0.542187511920929,0.18518517911434174,0.7583333551883698,0.39629629254341125],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.5057291984558105,0.37870368361473083,0.5677083693444729,0.4453703463077545],"occluded":false},{"timestamp":7.28,"box":[0.4869791865348816,0.4175925850868225,0.5453125238418579,0.4592592492699623],"occluded":false},{"timestamp":7.54,"box":[0.47708335518836975,0.43796294927597046,0.520312525331974,0.4851851724088192],"occluded":false},{"timestamp":7.8,"box":[0.45364585518836975,0.4462962746620178,0.4921875223517418,0.4953703470528126],"occluded":false},{"timestamp":8.06,"box":[0.4244791865348816,0.4592592418193817,0.4635416902601719,0.5120370201766491],"occluded":false},{"timestamp":8.32,"box":[0.39270836114883423,0.4740740656852722,0.4322916977107525,0.5296296216547489],"occluded":false},{"timestamp":8.58,"box":[0.34843751788139343,0.47870370745658875,0.39114585518836975,0.5314814858138561],"occluded":false},{"timestamp":8.84,"box":[0.2942708432674408,0.4870370328426361,0.3401041775941849,0.5462962910532951],"occluded":false},{"timestamp":9.1,"box":[0.25729167461395264,0.49444442987442017,0.3083333447575569,0.5537036880850792],"occluded":false},{"timestamp":9.36,"box":[0.20989584922790527,0.5138888955116272,0.26250001788139343,0.5750000067055225],"occluded":false},{"timestamp":9.620000000000001,"box":[0.19947917759418488,0.49444442987442017,0.2552083469927311,0.5574073940515518],"occluded":false},{"timestamp":9.88,"box":[0.2083333432674408,0.4731481373310089,0.26093751192092896,0.5398148000240326],"occluded":false},{"timestamp":10.14,"box":[0.1953125149011612,0.4564814865589142,0.2526041828095913,0.5231481492519379],"occluded":false},{"timestamp":10.4,"box":[0.17395834624767303,0.44722220301628113,0.2380208522081375,0.5129629448056221],"occluded":false},{"timestamp":10.66,"box":[0.15416666865348816,0.38425925374031067,0.21927084028720856,0.4574074000120163],"occluded":false},{"timestamp":10.92,"box":[0.14114584028720856,0.33703702688217163,0.2083333432674408,0.4138888791203499],"occluded":false},{"timestamp":11.18,"box":[0.1848958432674408,0.31018519401550293,0.2515625134110451,0.38055556267499924],"occluded":false},{"timestamp":11.44,"box":[0.20677083730697632,0.28148147463798523,0.2750000059604645,0.3490740656852722],"occluded":false},{"timestamp":11.700000000000001,"box":[0.22031250596046448,0.2611111104488373,0.2942708432674408,0.3314814791083336],"occluded":false},{"timestamp":11.96,"box":[0.10572917014360428,0.11574073880910873,0.3140625134110451,0.4027777835726738],"occluded":false},{"timestamp":12.22,"box":[0.3500000238418579,0.42222222685813904,0.395833358168602,0.4842592626810074],"occluded":false},{"timestamp":12.48,"box":[0.4442708492279053,0.49351850152015686,0.4817708507180214,0.5407407246530056],"occluded":false},{"timestamp":12.74,"box":[0.5119792222976685,0.45185184478759766,0.5302083902060986,0.506481472402811],"occluded":false},{"timestamp":13.0,"box":[0.5322917103767395,0.4277777671813965,0.5494792107492685,0.48148147016763687],"occluded":false},{"timestamp":13.26,"box":[0.5244792103767395,0.4592592418193817,0.5432292111217976,0.5222222059965134],"occluded":false},{"timestamp":13.52,"box":[0.5177083611488342,0.46296295523643494,0.5369791947305202,0.522222213447094],"occluded":false},{"timestamp":13.780000000000001,"box":[0.5151041746139526,0.46203702688217163,0.5343750081956387,0.5175925828516483],"occluded":false},{"timestamp":14.040000000000001,"box":[0.5234375,0.45277777314186096,0.544270833954215,0.504629623144865],"occluded":false},{"timestamp":14.3,"box":[0.5229166746139526,0.40740740299224854,0.5421875081956387,0.4574074037373066],"occluded":false},{"timestamp":14.56,"box":[0.515625,0.3898147940635681,0.5375000014901161,0.4462962746620178],"occluded":false},{"timestamp":14.82,"box":[0.5067708492279053,0.4037036895751953,0.5395833514630795,0.45277776196599007],"occluded":false},{"timestamp":15.08,"box":[0.48645836114883423,0.385185182094574,0.5338541977107525,0.43796296045184135],"occluded":false},{"timestamp":15.34,"box":[0.4713541865348816,0.3740740716457367,0.5223958566784859,0.42037036642432213],"occluded":false},{"timestamp":15.6,"box":[0.4661458432674408,0.3740740716457367,0.5208333469927311,0.43518518283963203],"occluded":false},{"timestamp":15.860000000000001,"box":[0.47187501192092896,0.2925925850868225,0.5250000134110451,0.3462962880730629],"occluded":false},{"timestamp":16.12,"box":[0.476041704416275,0.2518518567085266,0.5333333723247051,0.3074074126780033],"occluded":false},{"timestamp":16.38,"box":[0.49635419249534607,0.2944444417953491,0.5546875298023224,0.34166666492819786],"occluded":false},{"timestamp":16.64,"box":[0.48750001192092896,0.3166666626930237,0.5604166835546494,0.36481481045484543],"occluded":false},{"timestamp":16.9,"box":[0.4619791805744171,0.31481480598449707,0.5260416865348816,0.3722222112119198],"occluded":false},{"timestamp":17.16,"box":[0.44166669249534607,0.3462962806224823,0.5130208656191826,0.412962943315506],"occluded":false},{"timestamp":17.42,"box":[0.3437500298023224,0.31388887763023376,0.4135416969656944,0.3722222112119198],"occluded":false},{"timestamp":17.68,"box":[0.17135417461395264,0.2342592477798462,0.2640625163912773,0.29629628360271454],"occluded":false},{"timestamp":17.94,"box":[0.0,0.1111111119389534,0.07916667312383652,0.20370370149612427],"occluded":false}],"scores":[0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919]},{"noun_phrase":"brown dog","object_id":1,"observations":[{"timestamp":29.900000000000002,"box":[0.5093750357627869,0.7824074029922485,0.5453125387430191,0.8435185141861439],"occluded":false},{"timestamp":30.16,"box":[0.5119792222976685,0.7907407283782959,0.5510417260229588,0.8564814701676369],"occluded":false},{"timestamp":30.42,"box":[0.5114583373069763,0.7981481552124023,0.5546875074505806,0.8666666746139526],"occluded":false},{"timestamp":30.68,"box":[0.5145833492279053,0.8083333373069763,0.5541666857898235,0.8824074119329453],"occluded":false},{"timestamp":30.94,"box":[0.5130208730697632,0.8185185194015503,0.5567708760499954,0.8916666656732559],"occluded":false},{"timestamp":31.2,"box":[0.5104166865348816,0.8203703761100769,0.5619791895151138,0.8953703716397285],"occluded":false},{"timestamp":31.46,"box":[0.5135416984558105,0.8231481313705444,0.5671875327825546,0.902777761220932],"occluded":false},{"timestamp":31.720000000000002,"box":[0.510937511920929,0.8287037014961243,0.5656250156462193,0.9055555537343025],"occluded":false},{"timestamp":31.98,"box":[0.5104166865348816,0.8287037014961243,0.5630208551883698,0.9101851806044579],"occluded":false}],"scores":[0.8012422323226929,0.8012422323226929,0.8012422323226929,0.8012422323226929,0.8012422323226929,0.8012422323226929,0.8012422323226929,0.8012422323226929,0.8012422323226929]}],"tracklets":[{"noun_phrase":"brown dog","object_id":0,"presentation_id":1,"observations":[{"timestamp":0.26,"box":[0.20416668057441711,0.11944444477558136,0.24895834922790527,0.17777777835726738],"occluded":false},{"timestamp":0.52,"box":[0.23541668057441711,0.11203703284263611,0.28072918206453323,0.1657407358288765],"occluded":false},{"timestamp":0.78,"box":[0.2838541865348816,0.10370370000600815,0.32083335518836975,0.1509259231388569],"occluded":false},{"timestamp":1.04,"box":[0.3296875059604645,0.07222221791744232,0.35989584028720856,0.12129629030823708],"occluded":false},{"timestamp":1.3,"box":[0.3812500238418579,0.04814814776182175,0.40312502533197403,0.0981481485068798],"occluded":false},{"timestamp":1.56,"box":[0.43906253576278687,0.032407406717538834,0.4598958697170019,0.06851851567626],"occluded":false},{"timestamp":5.98,"box":[0.7026041746139526,0.031481482088565826,0.7723958417773247,0.06481481343507767],"occluded":false},{"timestamp":6.24,"box":[0.6427083611488342,0.14444443583488464,0.7026041969656944,0.19074073061347008],"occluded":false},{"timestamp":6.5,"box":[0.5880208611488342,0.2611111104488373,0.6463541984558105,0.316666666418314],"occluded":false},{"timestamp":6.76,"box":[0.542187511920929,0.18518517911434174,0.7583333551883698,0.39629629254341125],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.5057291984558105,0.37870368361473083,0.5677083693444729,0.4453703463077545],"occluded":false},{"timestamp":7.28,"box":[0.4869791865348816,0.4175925850868225,0.5453125238418579,0.4592592492699623],"occluded":false},{"timestamp":7.54,"box":[0.47708335518836975,0.43796294927597046,0.520312525331974,0.4851851724088192],"occluded":false},{"timestamp":7.8,"box":[0.45364585518836975,0.4462962746620178,0.4921875223517418,0.4953703470528126],"occluded":false},{"timestamp":8.06,"box":[0.4244791865348816,0.4592592418193817,0.4635416902601719,0.5120370201766491],"occluded":false},{"timestamp":8.32,"box":[0.39270836114883423,0.4740740656852722,0.4322916977107525,0.5296296216547489],"occluded":false},{"timestamp":8.58,"box":[0.34843751788139343,0.47870370745658875,0.39114585518836975,0.5314814858138561],"occluded":false},{"timestamp":8.84,"box":[0.2942708432674408,0.4870370328426361,0.3401041775941849,0.5462962910532951],"occluded":false},{"timestamp":9.1,"box":[0.25729167461395264,0.49444442987442017,0.3083333447575569,0.5537036880850792],"occluded":false},{"timestamp":9.36,"box":[0.20989584922790527,0.5138888955116272,0.26250001788139343,0.5750000067055225],"occluded":false},{"timestamp":9.620000000000001,"box":[0.19947917759418488,0.49444442987442017,0.2552083469927311,0.5574073940515518],"occluded":false},{"timestamp":9.88,"box":[0.2083333432674408,0.4731481373310089,0.26093751192092896,0.5398148000240326],"occluded":false},{"timestamp":10.14,"box":[0.1953125149011612,0.4564814865589142,0.2526041828095913,0.5231481492519379],"occluded":false},{"timestamp":10.4,"box":[0.17395834624767303,0.44722220301628113,0.2380208522081375,0.5129629448056221],"occluded":false},{"timestamp":10.66,"box":[0.15416666865348816,0.38425925374031067,0.21927084028720856,0.4574074000120163],"occluded":false},{"timestamp":10.92,"box":[0.14114584028720856,0.33703702688217163,0.2083333432674408,0.4138888791203499],"occluded":false},{"timestamp":11.18,"box":[0.1848958432674408,0.31018519401550293,0.2515625134110451,0.38055556267499924],"occluded":false},{"timestamp":11.44,"box":[0.20677083730697632,0.28148147463798523,0.2750000059604645,0.3490740656852722],"occluded":false},{"timestamp":11.700000000000001,"box":[0.22031250596046448,0.2611111104488373,0.2942708432674408,0.3314814791083336],"occluded":false},{"timestamp":11.96,"box":[0.10572917014360428,0.11574073880910873,0.3140625134110451,0.4027777835726738],"occluded":false},{"timestamp":12.22,"box":[0.3500000238418579,0.42222222685813904,0.395833358168602,0.4842592626810074],"occluded":false},{"timestamp":12.48,"box":[0.4442708492279053,0.49351850152015686,0.4817708507180214,0.5407407246530056],"occluded":false},{"timestamp":12.74,"box":[0.5119792222976685,0.45185184478759766,0.5302083902060986,0.506481472402811],"occluded":false},{"timestamp":13.0,"box":[0.5322917103767395,0.4277777671813965,0.5494792107492685,0.48148147016763687],"occluded":false},{"timestamp":13.26,"box":[0.5244792103767395,0.4592592418193817,0.5432292111217976,0.5222222059965134],"occluded":false},{"timestamp":13.52,"box":[0.5177083611488342,0.46296295523643494,0.5369791947305202,0.522222213447094],"occluded":false},{"timestamp":13.780000000000001,"box":[0.5151041746139526,0.46203702688217163,0.5343750081956387,0.5175925828516483],"occluded":false},{"timestamp":14.040000000000001,"box":[0.5234375,0.45277777314186096,0.544270833954215,0.504629623144865],"occluded":false},{"timestamp":14.3,"box":[0.5229166746139526,0.40740740299224854,0.5421875081956387,0.4574074037373066],"occluded":false},{"timestamp":14.56,"box":[0.515625,0.3898147940635681,0.5375000014901161,0.4462962746620178],"occluded":false},{"timestamp":14.82,"box":[0.5067708492279053,0.4037036895751953,0.5395833514630795,0.45277776196599007],"occluded":false},{"timestamp":15.08,"box":[0.48645836114883423,0.385185182094574,0.5338541977107525,0.43796296045184135],"occluded":false},{"timestamp":15.34,"box":[0.4713541865348816,0.3740740716457367,0.5223958566784859,0.42037036642432213],"occluded":false},{"timestamp":15.6,"box":[0.4661458432674408,0.3740740716457367,0.5208333469927311,0.43518518283963203],"occluded":false},{"timestamp":15.860000000000001,"box":[0.47187501192092896,0.2925925850868225,0.5250000134110451,0.3462962880730629],"occluded":false},{"timestamp":16.12,"box":[0.476041704416275,0.2518518567085266,0.5333333723247051,0.3074074126780033],"occluded":false},{"timestamp":16.38,"box":[0.49635419249534607,0.2944444417953491,0.5546875298023224,0.34166666492819786],"occluded":false},{"timestamp":16.64,"box":[0.48750001192092896,0.3166666626930237,0.5604166835546494,0.36481481045484543],"occluded":false},{"timestamp":16.9,"box":[0.4619791805744171,0.31481480598449707,0.5260416865348816,0.3722222112119198],"occluded":false},{"timestamp":17.16,"box":[0.44166669249534607,0.3462962806224823,0.5130208656191826,0.412962943315506],"occluded":false},{"timestamp":17.42,"box":[0.3437500298023224,0.31388887763023376,0.4135416969656944,0.3722222112119198],"occluded":false},{"timestamp":17.68,"box":[0.17135417461395264,0.2342592477798462,0.2640625163912773,0.29629628360271454],"occluded":false},{"timestamp":17.94,"box":[0.0,0.1111111119389534,0.07916667312383652,0.20370370149612427],"occluded":false}],"mean_score":0.8987342119216919},{"noun_phrase":"brown dog","object_id":1,"presentation_id":2,"observations":[{"timestamp":29.900000000000002,"box":[0.5093750357627869,0.7824074029922485,0.5453125387430191,0.8435185141861439],"occluded":false},{"timestamp":30.16,"box":[0.5119792222976685,0.7907407283782959,0.5510417260229588,0.8564814701676369],"occluded":false},{"timestamp":30.42,"box":[0.5114583373069763,0.7981481552124023,0.5546875074505806,0.8666666746139526],"occluded":false},{"timestamp":30.68,"box":[0.5145833492279053,0.8083333373069763,0.5541666857898235,0.8824074119329453],"occluded":false},{"timestamp":30.94,"box":[0.5130208730697632,0.8185185194015503,0.5567708760499954,0.8916666656732559],"occluded":false},{"timestamp":31.2,"box":[0.5104166865348816,0.8203703761100769,0.5619791895151138,0.8953703716397285],"occluded":false},{"timestamp":31.46,"box":[0.5135416984558105,0.8231481313705444,0.5671875327825546,0.902777761220932],"occluded":false},{"timestamp":31.720000000000002,"box":[0.510937511920929,0.8287037014961243,0.5656250156462193,0.9055555537343025],"occluded":false},{"timestamp":31.98,"box":[0.5104166865348816,0.8287037014961243,0.5630208551883698,0.9101851806044579],"occluded":false}],"mean_score":0.8012422323226929},{"noun_phrase":"black and white dog","object_id":0,"presentation_id":3,"observations":[{"timestamp":0.0,"box":[0.37291669845581055,0.4851851761341095,0.6661458909511566,0.8083333075046539],"occluded":false},{"timestamp":0.26,"box":[0.33385419845581055,0.470370352268219,0.6234375536441803,0.8083333075046539],"occluded":false},{"timestamp":0.52,"box":[0.2953125238418579,0.45185184478759766,0.5906250476837158,0.8620370328426361],"occluded":false},{"timestamp":0.78,"box":[0.2526041865348816,0.43981480598449707,0.572916716337204,0.852777749300003],"occluded":false},{"timestamp":1.04,"box":[0.21562501788139343,0.45370370149612427,0.5661458671092987,0.8620370328426361],"occluded":false},{"timestamp":1.3,"box":[0.19479167461395264,0.44722220301628113,0.5682291984558105,0.8481481075286865],"occluded":false},{"timestamp":1.56,"box":[0.1848958432674408,0.43703702092170715,0.5489583611488342,0.8175925612449646],"occluded":false},{"timestamp":1.82,"box":[0.16718751192092896,0.4120370149612427,0.5359375178813934,0.8324073851108551],"occluded":false},{"timestamp":2.08,"box":[0.13593751192092896,0.4259259104728699,0.5302083790302277,0.8175925612449646],"occluded":false},{"timestamp":2.34,"box":[0.12031250447034836,0.4157407283782959,0.5250000283122063,0.8564814627170563],"occluded":false},{"timestamp":2.6,"box":[0.1640625149011612,0.40833333134651184,0.5609375387430191,0.8148148059844971],"occluded":false},{"timestamp":2.86,"box":[0.24947917461395264,0.3305555582046509,0.6994791924953461,0.9990740418434143],"occluded":false},{"timestamp":3.12,"box":[0.3281250298023224,0.3472222089767456,0.744791716337204,0.7435185015201569],"occluded":false},{"timestamp":3.38,"box":[0.3921875059604645,0.43611109256744385,0.7968750298023224,0.8120370209217072],"occluded":false},{"timestamp":3.64,"box":[0.40677085518836975,0.4861111044883728,0.7708333730697632,0.8527777791023254],"occluded":false},{"timestamp":3.9,"box":[0.39531251788139343,0.5175926089286804,0.7067708671092987,0.8666666746139526],"occluded":false},{"timestamp":4.16,"box":[0.39635419845581055,0.5046296119689941,0.6640625596046448,0.8425925672054291],"occluded":false},{"timestamp":4.42,"box":[0.4208333492279053,0.4888888895511627,0.6505208611488342,0.8638888895511627],"occluded":false},{"timestamp":4.68,"box":[0.43541669845581055,0.47592592239379883,0.6348958760499954,0.8759259283542633],"occluded":false},{"timestamp":4.94,"box":[0.4166666865348816,0.46759259700775146,0.602083370089531,0.8935185074806213],"occluded":false},{"timestamp":5.2,"box":[0.40000003576278687,0.44814813137054443,0.5354167073965073,0.912962943315506],"occluded":false},{"timestamp":5.46,"box":[0.3348958492279053,0.5046296119689941,0.49114586412906647,0.8666666448116302],"occluded":false},{"timestamp":15.6,"box":[0.46562501788139343,0.375,0.5203125216066837,0.43518518283963203],"occluded":false}],"mean_score":0.9452055096626282}],"dropped":{},"fates":[{"noun_phrase":"black and white dog","object_id":0,"frames":23,"mean_score":0.9452055096626282,"mean_area":0.12369687313629037,"outcome":"kept","presentation_id":3,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"brown dog","object_id":0,"frames":53,"mean_score":0.8987342119216919,"mean_area":0.004721658315924237,"outcome":"kept","presentation_id":1,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"brown dog","object_id":1,"frames":9,"mean_score":0.8012422323226929,"mean_area":0.0033873994374345583,"outcome":"kept","presentation_id":2,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null}],"gates":{"min_track_frames":3,"min_mean_score":0.5,"merge_iou":0.8,"max_tracklets":12}}},"render":{"inputs":{"tracklets":3,"as_video":true,"fps":1.0,"max_frames":30},"outputs":{"media":"/home/zzk/data/video_bbox/vlm_gemini3.5_lite/renders/--cif7Nv6BI_chunk_003_scene_023_choices.mp4","media_bytes":11580219,"frames":[{"index":1,"timestamp":0.0,"file":"--cif7Nv6BI_chunk_003_scene_023_choices.mp4"},{"index":2,"timestamp":1.04,"file":"--cif7Nv6BI_chunk_003_scene_023_choices.mp4"},{"index":3,"timestamp":2.08,"file":"--cif7Nv6BI_chunk_003_scene_023_choices.mp4"},{"index":4,"timestamp":3.38,"file":"--cif7Nv6BI_chunk_003_scene_023_choices.mp4"},{"index":5,"timestamp":4.42,"file":"--cif7Nv6BI_chunk_003_scene_023_choices.mp4"},{"index":6,"timestamp":5.46,"file":"--cif7Nv6BI_chunk_003_scene_023_choices.mp4"},{"index":7,"timestamp":6.5,"file":"--cif7Nv6BI_chunk_003_scene_023_choices.mp4"},{"index":8,"timestamp":7.8,"file":"--cif7Nv6BI_chunk_003_scene_023_choices.mp4"},{"index":9,"timestamp":8.84,"file":"--cif7Nv6BI_chunk_003_scene_023_choices.mp4"},{"index":10,"timestamp":9.88,"file":"--cif7Nv6BI_chunk_003_scene_023_choices.mp4"},{"index":11,"timestamp":10.92,"file":"--cif7Nv6BI_chunk_003_scene_023_choices.mp4"},{"index":12,"timestamp":12.22,"file":"--cif7Nv6BI_chunk_003_scene_023_choices.mp4"},{"index":13,"timestamp":13.26,"file":"--cif7Nv6BI_chunk_003_scene_023_choices.mp4"},{"index":14,"timestamp":14.3,"file":"--cif7Nv6BI_chunk_003_scene_023_choices.mp4"},{"index":15,"timestamp":15.34,"file":"--cif7Nv6BI_chunk_003_scene_023_choices.mp4"},{"index":16,"timestamp":16.64,"file":"--cif7Nv6BI_chunk_003_scene_023_choices.mp4"},{"index":17,"timestamp":17.68,"file":"--cif7Nv6BI_chunk_003_scene_023_choices.mp4"},{"index":18,"timestamp":18.72,"file":"--cif7Nv6BI_chunk_003_scene_023_choices.mp4"},{"index":19,"timestamp":19.76,"file":"--cif7Nv6BI_chunk_003_scene_023_choices.mp4"},{"index":20,"timestamp":21.06,"file":"--cif7Nv6BI_chunk_003_scene_023_choices.mp4"},{"index":21,"timestamp":22.1,"file":"--cif7Nv6BI_chunk_003_scene_023_choices.mp4"},{"index":22,"timestamp":23.14,"file":"--cif7Nv6BI_chunk_003_scene_023_choices.mp4"},{"index":23,"timestamp":24.18,"file":"--cif7Nv6BI_chunk_003_scene_023_choices.mp4"},{"index":24,"timestamp":25.48,"file":"--cif7Nv6BI_chunk_003_scene_023_choices.mp4"},{"index":25,"timestamp":26.52,"file":"--cif7Nv6BI_chunk_003_scene_023_choices.mp4"},{"index":26,"timestamp":27.560000000000002,"file":"--cif7Nv6BI_chunk_003_scene_023_choices.mp4"},{"index":27,"timestamp":28.6,"file":"--cif7Nv6BI_chunk_003_scene_023_choices.mp4"},{"index":28,"timestamp":29.900000000000002,"file":"--cif7Nv6BI_chunk_003_scene_023_choices.mp4"},{"index":29,"timestamp":30.94,"file":"--cif7Nv6BI_chunk_003_scene_023_choices.mp4"},{"index":30,"timestamp":31.98,"file":"--cif7Nv6BI_chunk_003_scene_023_choices.mp4"}]}},"select":{"inputs":{"model":"gemini-3.5-flash-lite","temperature":0.0,"as_video":true,"media_bytes":11580219,"media":["--cif7Nv6BI_chunk_003_scene_023_choices.mp4"],"choice_frames":30,"system_prompt":"You are labelling video data for referring multi-object tracking.\n\nYou will see a numbered sequence of frames from one clip, in time order. Each frame has its own number and timestamp printed in the top-left corner. Candidate objects have been detected for you and drawn on the frames as coloured boxes, each labelled with a number. The same object keeps the same number and the same colour on every frame. Below each frame is a strip of enlarged crops of those same numbered boxes, so you can identify a small or distant object.\n\nFor each description you are given, answer two questions:\n\n1. MATCHING_IDS — which of the numbered objects the description refers to. List every number that matches, or an EMPTY LIST if none of them do. An empty list is a correct and expected answer: some descriptions are deliberately about something that is not in this clip. Do not pick a number just to have picked one, and do not pick a number for an object that merely resembles the description.\n\n2. START_FRAME and END_FRAME — the frame numbers between which the described state actually holds. If the description is \"the car that's turning left\", the interval is the frames in which that car is turning, NOT every frame in which the car is visible. Use the frame numbers printed on the frames. START_FRAME must not be greater than END_FRAME. When MATCHING_IDS is empty, give the full range (the first and last frame numbers); it is ignored.\n\n3. CONFIDENCE — how sure you are of the MATCHING_IDS answer, from 0.0 to 1.0. Be honest: a low number costs the example nothing, and an overconfident wrong answer becomes a wrong label.\n\nRules:\n- Only use numbers that are actually drawn on the frames.\n- Never output coordinates, box values, or pixel positions.\n- Judge motion by comparing the object's position across frames, not by guessing from one frame.\n","user_prompt":"The video has 30 frames, numbered 1 to 30. Each frame is one moment of the clip and has its number burned into the top left corner.\n\nAnswer for each of the following descriptions, in the order given:\n1. the black and white dog walking on the stone pavement\n2. the brown dog walking up the stone stairs in the background","questions":["the black and white dog walking on the stone pavement","the brown dog walking up the stone stairs in the background"]},"outputs":{"raw_response":"{\n  \"answers\": [\n    {\n      \"description_index\": 1,\n      \"matching_ids\": [\n        3\n      ],\n      \"start_frame\": 1,\n      \"end_frame\": 6,\n      \"confidence\": 1.0\n    },\n    {\n      \"description_index\": 2,\n      \"matching_ids\": [\n        1\n      ],\n      \"start_frame\": 7,\n      \"end_frame\": 17,\n      \"confidence\": 1.0\n    }\n  ]\n}","raw_answers":[{"description_index":1,"matching_ids":[3],"start_frame":1,"end_frame":6,"confidence":1.0},{"description_index":2,"matching_ids":[1],"start_frame":7,"end_frame":17,"confidence":1.0}],"answers":[{"expression":"the black and white dog walking on the stone pavement","matching_presentation_ids":[3],"start_choice_frame":1,"end_choice_frame":6,"confidence":1.0,"self_reported_confidence":1.0},{"expression":"the brown dog walking up the stone stairs in the background","matching_presentation_ids":[1],"start_choice_frame":7,"end_choice_frame":17,"confidence":1.0,"self_reported_confidence":1.0}]}},"verify":{"inputs":{"expressions":2,"tracklets":3,"selections":2,"gates":"interval, count, unknown ids, confidence, then geometry (motion)"},"outputs":{"kept":[{"expression":"the black and white dog walking on the stone pavement","kind":"motion","is_negative":false,"tracks":[{"timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46],"boxes":[[0.37291669845581055,0.4851851761341095,0.6661458909511566,0.8083333075046539],[0.33385419845581055,0.470370352268219,0.6234375536441803,0.8083333075046539],[0.2953125238418579,0.45185184478759766,0.5906250476837158,0.8620370328426361],[0.2526041865348816,0.43981480598449707,0.572916716337204,0.852777749300003],[0.21562501788139343,0.45370370149612427,0.5661458671092987,0.8620370328426361],[0.19479167461395264,0.44722220301628113,0.5682291984558105,0.8481481075286865],[0.1848958432674408,0.43703702092170715,0.5489583611488342,0.8175925612449646],[0.16718751192092896,0.4120370149612427,0.5359375178813934,0.8324073851108551],[0.13593751192092896,0.4259259104728699,0.5302083790302277,0.8175925612449646],[0.12031250447034836,0.4157407283782959,0.5250000283122063,0.8564814627170563],[0.1640625149011612,0.40833333134651184,0.5609375387430191,0.8148148059844971],[0.24947917461395264,0.3305555582046509,0.6994791924953461,0.9990740418434143],[0.3281250298023224,0.3472222089767456,0.744791716337204,0.7435185015201569],[0.3921875059604645,0.43611109256744385,0.7968750298023224,0.8120370209217072],[0.40677085518836975,0.4861111044883728,0.7708333730697632,0.8527777791023254],[0.39531251788139343,0.5175926089286804,0.7067708671092987,0.8666666746139526],[0.39635419845581055,0.5046296119689941,0.6640625596046448,0.8425925672054291],[0.4208333492279053,0.4888888895511627,0.6505208611488342,0.8638888895511627],[0.43541669845581055,0.47592592239379883,0.6348958760499954,0.8759259283542633],[0.4166666865348816,0.46759259700775146,0.602083370089531,0.8935185074806213],[0.40000003576278687,0.44814813137054443,0.5354167073965073,0.912962943315506],[0.3348958492279053,0.5046296119689941,0.49114586412906647,0.8666666448116302]]}]},{"expression":"the brown dog walking up the stone stairs in the background","kind":"motion","is_negative":false,"tracks":[{"timestamps":[6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88,10.14,10.4,10.66,10.92,11.18,11.44,11.700000000000001,11.96,12.22,12.48,12.74,13.0,13.26,13.52,13.780000000000001,14.040000000000001,14.3,14.56,14.82,15.08,15.34,15.6,15.860000000000001,16.12,16.38,16.64,16.9,17.16,17.42,17.68],"boxes":[[0.5880208611488342,0.2611111104488373,0.6463541984558105,0.316666666418314],[0.542187511920929,0.18518517911434174,0.7583333551883698,0.39629629254341125],[0.5057291984558105,0.37870368361473083,0.5677083693444729,0.4453703463077545],[0.4869791865348816,0.4175925850868225,0.5453125238418579,0.4592592492699623],[0.47708335518836975,0.43796294927597046,0.520312525331974,0.4851851724088192],[0.45364585518836975,0.4462962746620178,0.4921875223517418,0.4953703470528126],[0.4244791865348816,0.4592592418193817,0.4635416902601719,0.5120370201766491],[0.39270836114883423,0.4740740656852722,0.4322916977107525,0.5296296216547489],[0.34843751788139343,0.47870370745658875,0.39114585518836975,0.5314814858138561],[0.2942708432674408,0.4870370328426361,0.3401041775941849,0.5462962910532951],[0.25729167461395264,0.49444442987442017,0.3083333447575569,0.5537036880850792],[0.20989584922790527,0.5138888955116272,0.26250001788139343,0.5750000067055225],[0.19947917759418488,0.49444442987442017,0.2552083469927311,0.5574073940515518],[0.2083333432674408,0.4731481373310089,0.26093751192092896,0.5398148000240326],[0.1953125149011612,0.4564814865589142,0.2526041828095913,0.5231481492519379],[0.17395834624767303,0.44722220301628113,0.2380208522081375,0.5129629448056221],[0.15416666865348816,0.38425925374031067,0.21927084028720856,0.4574074000120163],[0.14114584028720856,0.33703702688217163,0.2083333432674408,0.4138888791203499],[0.1848958432674408,0.31018519401550293,0.2515625134110451,0.38055556267499924],[0.20677083730697632,0.28148147463798523,0.2750000059604645,0.3490740656852722],[0.22031250596046448,0.2611111104488373,0.2942708432674408,0.3314814791083336],[0.10572917014360428,0.11574073880910873,0.3140625134110451,0.4027777835726738],[0.3500000238418579,0.42222222685813904,0.395833358168602,0.4842592626810074],[0.4442708492279053,0.49351850152015686,0.4817708507180214,0.5407407246530056],[0.5119792222976685,0.45185184478759766,0.5302083902060986,0.506481472402811],[0.5322917103767395,0.4277777671813965,0.5494792107492685,0.48148147016763687],[0.5244792103767395,0.4592592418193817,0.5432292111217976,0.5222222059965134],[0.5177083611488342,0.46296295523643494,0.5369791947305202,0.522222213447094],[0.5151041746139526,0.46203702688217163,0.5343750081956387,0.5175925828516483],[0.5234375,0.45277777314186096,0.544270833954215,0.504629623144865],[0.5229166746139526,0.40740740299224854,0.5421875081956387,0.4574074037373066],[0.515625,0.3898147940635681,0.5375000014901161,0.4462962746620178],[0.5067708492279053,0.4037036895751953,0.5395833514630795,0.45277776196599007],[0.48645836114883423,0.385185182094574,0.5338541977107525,0.43796296045184135],[0.4713541865348816,0.3740740716457367,0.5223958566784859,0.42037036642432213],[0.4661458432674408,0.3740740716457367,0.5208333469927311,0.43518518283963203],[0.47187501192092896,0.2925925850868225,0.5250000134110451,0.3462962880730629],[0.476041704416275,0.2518518567085266,0.5333333723247051,0.3074074126780033],[0.49635419249534607,0.2944444417953491,0.5546875298023224,0.34166666492819786],[0.48750001192092896,0.3166666626930237,0.5604166835546494,0.36481481045484543],[0.4619791805744171,0.31481480598449707,0.5260416865348816,0.3722222112119198],[0.44166669249534607,0.3462962806224823,0.5130208656191826,0.412962943315506],[0.3437500298023224,0.31388887763023376,0.4135416969656944,0.3722222112119198],[0.17135417461395264,0.2342592477798462,0.2640625163912773,0.29629628360271454]]}]}],"rejected":{},"rejection_counts":{},"row":{"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_003_scene_023.mp4","scene_key":"--cif7Nv6BI_chunk_003_scene_023","category":"test_vehicles","height":1080,"width":1920,"fps":50.0,"num_frames":1600,"duration":32.0,"rmot_expressions":[{"expression":"the black and white dog walking on the stone pavement","kind":"motion","is_negative":false,"tracks":[{"timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46],"boxes":[[0.37291669845581055,0.4851851761341095,0.6661458909511566,0.8083333075046539],[0.33385419845581055,0.470370352268219,0.6234375536441803,0.8083333075046539],[0.2953125238418579,0.45185184478759766,0.5906250476837158,0.8620370328426361],[0.2526041865348816,0.43981480598449707,0.572916716337204,0.852777749300003],[0.21562501788139343,0.45370370149612427,0.5661458671092987,0.8620370328426361],[0.19479167461395264,0.44722220301628113,0.5682291984558105,0.8481481075286865],[0.1848958432674408,0.43703702092170715,0.5489583611488342,0.8175925612449646],[0.16718751192092896,0.4120370149612427,0.5359375178813934,0.8324073851108551],[0.13593751192092896,0.4259259104728699,0.5302083790302277,0.8175925612449646],[0.12031250447034836,0.4157407283782959,0.5250000283122063,0.8564814627170563],[0.1640625149011612,0.40833333134651184,0.5609375387430191,0.8148148059844971],[0.24947917461395264,0.3305555582046509,0.6994791924953461,0.9990740418434143],[0.3281250298023224,0.3472222089767456,0.744791716337204,0.7435185015201569],[0.3921875059604645,0.43611109256744385,0.7968750298023224,0.8120370209217072],[0.40677085518836975,0.4861111044883728,0.7708333730697632,0.8527777791023254],[0.39531251788139343,0.5175926089286804,0.7067708671092987,0.8666666746139526],[0.39635419845581055,0.5046296119689941,0.6640625596046448,0.8425925672054291],[0.4208333492279053,0.4888888895511627,0.6505208611488342,0.8638888895511627],[0.43541669845581055,0.47592592239379883,0.6348958760499954,0.8759259283542633],[0.4166666865348816,0.46759259700775146,0.602083370089531,0.8935185074806213],[0.40000003576278687,0.44814813137054443,0.5354167073965073,0.912962943315506],[0.3348958492279053,0.5046296119689941,0.49114586412906647,0.8666666448116302]]}]},{"expression":"the brown dog walking up the stone stairs in the background","kind":"motion","is_negative":false,"tracks":[{"timestamps":[6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88,10.14,10.4,10.66,10.92,11.18,11.44,11.700000000000001,11.96,12.22,12.48,12.74,13.0,13.26,13.52,13.780000000000001,14.040000000000001,14.3,14.56,14.82,15.08,15.34,15.6,15.860000000000001,16.12,16.38,16.64,16.9,17.16,17.42,17.68],"boxes":[[0.5880208611488342,0.2611111104488373,0.6463541984558105,0.316666666418314],[0.542187511920929,0.18518517911434174,0.7583333551883698,0.39629629254341125],[0.5057291984558105,0.37870368361473083,0.5677083693444729,0.4453703463077545],[0.4869791865348816,0.4175925850868225,0.5453125238418579,0.4592592492699623],[0.47708335518836975,0.43796294927597046,0.520312525331974,0.4851851724088192],[0.45364585518836975,0.4462962746620178,0.4921875223517418,0.4953703470528126],[0.4244791865348816,0.4592592418193817,0.4635416902601719,0.5120370201766491],[0.39270836114883423,0.4740740656852722,0.4322916977107525,0.5296296216547489],[0.34843751788139343,0.47870370745658875,0.39114585518836975,0.5314814858138561],[0.2942708432674408,0.4870370328426361,0.3401041775941849,0.5462962910532951],[0.25729167461395264,0.49444442987442017,0.3083333447575569,0.5537036880850792],[0.20989584922790527,0.5138888955116272,0.26250001788139343,0.5750000067055225],[0.19947917759418488,0.49444442987442017,0.2552083469927311,0.5574073940515518],[0.2083333432674408,0.4731481373310089,0.26093751192092896,0.5398148000240326],[0.1953125149011612,0.4564814865589142,0.2526041828095913,0.5231481492519379],[0.17395834624767303,0.44722220301628113,0.2380208522081375,0.5129629448056221],[0.15416666865348816,0.38425925374031067,0.21927084028720856,0.4574074000120163],[0.14114584028720856,0.33703702688217163,0.2083333432674408,0.4138888791203499],[0.1848958432674408,0.31018519401550293,0.2515625134110451,0.38055556267499924],[0.20677083730697632,0.28148147463798523,0.2750000059604645,0.3490740656852722],[0.22031250596046448,0.2611111104488373,0.2942708432674408,0.3314814791083336],[0.10572917014360428,0.11574073880910873,0.3140625134110451,0.4027777835726738],[0.3500000238418579,0.42222222685813904,0.395833358168602,0.4842592626810074],[0.4442708492279053,0.49351850152015686,0.4817708507180214,0.5407407246530056],[0.5119792222976685,0.45185184478759766,0.5302083902060986,0.506481472402811],[0.5322917103767395,0.4277777671813965,0.5494792107492685,0.48148147016763687],[0.5244792103767395,0.4592592418193817,0.5432292111217976,0.5222222059965134],[0.5177083611488342,0.46296295523643494,0.5369791947305202,0.522222213447094],[0.5151041746139526,0.46203702688217163,0.5343750081956387,0.5175925828516483],[0.5234375,0.45277777314186096,0.544270833954215,0.504629623144865],[0.5229166746139526,0.40740740299224854,0.5421875081956387,0.4574074037373066],[0.515625,0.3898147940635681,0.5375000014901161,0.4462962746620178],[0.5067708492279053,0.4037036895751953,0.5395833514630795,0.45277776196599007],[0.48645836114883423,0.385185182094574,0.5338541977107525,0.43796296045184135],[0.4713541865348816,0.3740740716457367,0.5223958566784859,0.42037036642432213],[0.4661458432674408,0.3740740716457367,0.5208333469927311,0.43518518283963203],[0.47187501192092896,0.2925925850868225,0.5250000134110451,0.3462962880730629],[0.476041704416275,0.2518518567085266,0.5333333723247051,0.3074074126780033],[0.49635419249534607,0.2944444417953491,0.5546875298023224,0.34166666492819786],[0.48750001192092896,0.3166666626930237,0.5604166835546494,0.36481481045484543],[0.4619791805744171,0.31481480598449707,0.5260416865348816,0.3722222112119198],[0.44166669249534607,0.3462962806224823,0.5130208656191826,0.412962943315506],[0.3437500298023224,0.31388887763023376,0.4135416969656944,0.3722222112119198],[0.17135417461395264,0.2342592477798462,0.2640625163912773,0.29629628360271454]]}]}],"rmot_expression_count":2,"rmot_negative_count":0,"rmot_motion_count":2}}}}},{"index":2,"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_000_scene_044.mp4","status":"complete","steps":{"decode":{"inputs":{"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_000_scene_044.mp4","sample_fps":4.0},"outputs":{"width":1920,"height":1080,"source_fps":50.0,"duration_seconds":10.32,"frames_kept":40,"blank_frames_dropped":0,"timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88,10.14],"error":""}},"propose":{"inputs":{"model":"gemini-3.5-flash-lite","temperature":0.4,"video_bytes":12411656,"system_prompt":"You are annotating video clips to build training data for referring multi-object tracking: given a video and a natural-language description, a model must output the boxes of every object matching that description, over time.\n\nYour job is to write the descriptions. You do NOT output coordinates, boxes, pixel positions, or any numbers about where things are. A separate detector finds the objects.\n\nFor every clip you produce a list of items. Each item has two texts that must NOT be the same:\n\nEXPRESSION — what a person would say to single out the object or objects. It may describe motion, temporal order, interaction with other objects, and spatial relations. It should read like something a person would actually say.\n  Good: \"the black car that's turning left\"\n  Good: \"the cyclist who overtakes the bus\"\n  Good: \"the two people who walk out of frame on the right\"\n\nNOUN_PHRASE — the same object as a plain, timeless visual category with its visible attributes, and NOTHING else. The detector is prompted with this text and looks at ONE frame at a time, so anything about time, order, or action is useless to it and makes it fail.\n  Allowed: category nouns, colours, materials, clothing, size, type\n  Forbidden: verbs and participles of action (\"turning\", \"walks\", \"is moving\"), temporal words (\"then\", \"starts\", \"after\", \"first\", \"last\"), relative clauses (\"that ...\", \"which ...\", \"who ...\"), ordinals and rank words (\"first\", \"leftmost\", \"nearest\"), and anything over six words.\n  \"the black car that's turning left\" -> \"black car\"\n  \"the cyclist who overtakes the bus\" -> \"cyclist\"\n  \"the two people who walk out of frame on the right\" -> \"person\"\n\nKIND — what makes the expression hard: \"motion\", \"appearance\", \"spatial\", or \"interaction\". Aim for about 35% \"motion\" items: motion is the part a single-frame model cannot already solve, so it is the most valuable, but a list of only motion items is not usable training data either.\n\nEXPECTED_COUNT — how many distinct objects in the clip match the EXPRESSION. Count carefully; this is checked against the detector's answer, and a wrong count discards the item.\n\nIS_NEGATIVE — at most ONE item may be a negative, and only when you also write at least one non-negative item. A negative is an expression that recombines attributes that ARE present in the clip into a whole that is NOT. If the only black car is moving, \"the parked black car\" is a good negative. Its NOUN_PHRASE must still be the category that IS present (\"black car\"), because that is what makes the absence checkable rather than merely unverified. A negative's EXPECTED_COUNT is 0.\n\nRules that discard an item, so follow them exactly:\n- Never write coordinates, box values, pixel positions, or percentages of the frame.\n- Never let a NOUN_PHRASE contain a verb of action, a temporal word, a relative clause, or an ordinal.\n- Never write an EXPRESSION whose object is not visible in the clip, unless it is marked IS_NEGATIVE.\n- Prefer objects that are clearly visible and identifiable over background clutter.\n- Do not describe the camera or the shot; describe objects in the scene.\n","user_prompt":"Watch this clip and write between 1 and 3 items following the rules. Cover different objects rather than restating one object several ways, and include any negative in the same list.\n"},"outputs":{"raw_response":"{\n  \"items\": [\n    {\n      \"expression\": \"the white car parked along the stone building\",\n      \"noun_phrase\": \"white car\",\n      \"kind\": \"appearance\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the grey SUV parked next to the white car\",\n      \"noun_phrase\": \"grey SUV\",\n      \"kind\": \"spatial\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the red car parked on the street\",\n      \"noun_phrase\": \"red car\",\n      \"kind\": \"appearance\",\n      \"expected_count\": 0,\n      \"is_negative\": true\n    }\n  ]\n}","raw_items":[{"expression":"the white car parked along the stone building","noun_phrase":"white car","kind":"appearance","expected_count":1,"is_negative":false},{"expression":"the grey SUV parked next to the white car","noun_phrase":"grey SUV","kind":"spatial","expected_count":1,"is_negative":false},{"expression":"the red car parked on the street","noun_phrase":"red car","kind":"appearance","expected_count":0,"is_negative":true}],"accepted":[{"expression":"the white car parked along the stone building","noun_phrase":"white car","kind":"appearance","expected_count":1,"is_negative":false},{"expression":"the grey SUV parked next to the white car","noun_phrase":"grey SUV","kind":"spatial","expected_count":1,"is_negative":false},{"expression":"the red car parked on the street","noun_phrase":"red car","kind":"appearance","expected_count":0,"is_negative":true}],"gated":[]}},"track":{"inputs":{"tracker":"Sam31MultiplexTracker","model_id":"/mnt/ceph3/ec/jacky/sam3.1","noun_phrases":["grey SUV","red car","white car"],"expressions_in":3,"frames":40,"sample_fps":4.0,"frame_size":"1920x1080","timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88,10.14],"session":"one session over 40 JPEG frames; then per phrase: reset_session, add_prompt(text) on frame 0, propagate_in_video","passes":3},"outputs":{"raw_tracks":[{"noun_phrase":"grey SUV","object_id":0,"observations":[{"timestamp":0.0,"box":[0.9312500357627869,0.4259259104728699,0.999479204416275,0.7833333015441895],"occluded":false},{"timestamp":0.26,"box":[0.8828125596046448,0.4185185134410858,0.9994792342185974,0.787962943315506],"occluded":false},{"timestamp":0.52,"box":[0.8479167222976685,0.4175925850868225,0.9994792342185974,0.817592591047287],"occluded":false},{"timestamp":0.78,"box":[0.8796875476837158,0.4444444477558136,0.9994792193174362,0.8305555582046509],"occluded":false}],"scores":[0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295]},{"noun_phrase":"grey SUV","object_id":1,"observations":[{"timestamp":4.16,"box":[0.0,0.35462963581085205,0.0598958358168602,0.5009259283542633],"occluded":false},{"timestamp":4.42,"box":[0.01875000074505806,0.39814814925193787,0.23072918131947517,0.5601851791143417],"occluded":false},{"timestamp":4.68,"box":[0.2291666716337204,0.41018518805503845,0.35364584624767303,0.5453703701496124],"occluded":false},{"timestamp":4.94,"box":[0.32239586114883423,0.46759259700775146,0.43385419994592667,0.5916666686534882],"occluded":false},{"timestamp":5.2,"box":[0.3906250298023224,0.49351850152015686,0.4906250387430191,0.605555534362793],"occluded":false},{"timestamp":5.46,"box":[0.42239585518836975,0.5055555701255798,0.5161458626389503,0.6111111268401146],"occluded":false},{"timestamp":5.72,"box":[0.43541669845581055,0.4972222149372101,0.5354167073965073,0.6027777716517448],"occluded":false},{"timestamp":5.98,"box":[0.47083336114883423,0.5111110806465149,0.5645833685994148,0.6157407090067863],"occluded":false},{"timestamp":6.24,"box":[0.5067708492279053,0.5425925850868225,0.6026041880249977,0.6509259194135666],"occluded":false},{"timestamp":6.5,"box":[0.5307291746139526,0.5574073791503906,0.627604179084301,0.6648147851228714],"occluded":false},{"timestamp":6.76,"box":[0.5411458611488342,0.5675925612449646,0.6406250298023224,0.6777777448296547],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.5505208373069763,0.5814814567565918,0.651562511920929,0.6935184895992279],"occluded":false},{"timestamp":7.28,"box":[0.5921875238418579,0.5629629492759705,0.6979166939854622,0.6787036880850792],"occluded":false},{"timestamp":7.54,"box":[0.6463541984558105,0.5796296000480652,0.7567708715796471,0.6990740448236465],"occluded":false},{"timestamp":7.8,"box":[0.7010416984558105,0.550000011920929,0.8218750357627869,0.6768518686294556],"occluded":false},{"timestamp":8.06,"box":[0.7562500238418579,0.5148147940635681,0.8906250298023224,0.6499999761581421],"occluded":false},{"timestamp":8.32,"box":[0.8125000596046448,0.4861111044883728,0.965625062584877,0.6305555403232574],"occluded":false},{"timestamp":8.58,"box":[0.8791667222976685,0.4611110985279083,0.9994792267680168,0.6166666448116302],"occluded":false},{"timestamp":10.14,"box":[0.9187500476837158,0.4870370328426361,0.9994792193174362,0.6259259134531021],"occluded":false}],"scores":[0.8439024686813354,0.8439024686813354,0.8439024686813354,0.8439024686813354,0.8439024686813354,0.8439024686813354,0.8439024686813354,0.8439024686813354,0.8439024686813354,0.8439024686813354,0.8439024686813354,0.8439024686813354,0.8439024686813354,0.8439024686813354,0.8439024686813354,0.8439024686813354,0.8439024686813354,0.8439024686813354,0.8439024686813354]},{"noun_phrase":"red car","object_id":0,"observations":[{"timestamp":3.12,"box":[0.02916666865348816,0.6435185074806213,0.19427084922790527,0.8333333134651184],"occluded":false},{"timestamp":3.38,"box":[0.0390625037252903,0.6499999761581421,0.20520834997296333,0.8425925672054291],"occluded":false},{"timestamp":3.64,"box":[0.1093750074505806,0.6333333253860474,0.2645833417773247,0.8120370209217072],"occluded":false},{"timestamp":3.9,"box":[0.20208334922790527,0.6509259343147278,0.3421875238418579,0.817592591047287],"occluded":false},{"timestamp":4.16,"box":[0.35989585518836975,0.6527777910232544,0.4843750298023224,0.8064814954996109],"occluded":false},{"timestamp":4.42,"box":[0.4947916865348816,0.6527777910232544,0.622395858168602,0.8046296387910843],"occluded":false},{"timestamp":4.68,"box":[0.6088541746139526,0.6425926089286804,0.7427083551883698,0.7981481552124023],"occluded":false},{"timestamp":4.94,"box":[0.6765625476837158,0.6814814805984497,0.8208333849906921,0.8388888835906982],"occluded":false},{"timestamp":5.2,"box":[0.7192708849906921,0.6824073791503906,0.8609375655651093,0.8351851552724838],"occluded":false},{"timestamp":5.46,"box":[0.7432292103767395,0.6953703761100769,0.8906250447034836,0.8527777791023254],"occluded":false},{"timestamp":5.72,"box":[0.7645833492279053,0.6925925612449646,0.9239583611488342,0.8574073761701584],"occluded":false},{"timestamp":5.98,"box":[0.8072916865348816,0.7138888835906982,0.9885416924953461,0.8981481343507767],"occluded":false}],"scores":[0.6604774594306946,0.6604774594306946,0.6604774594306946,0.6604774594306946,0.6604774594306946,0.6604774594306946,0.6604774594306946,0.6604774594306946,0.6604774594306946,0.6604774594306946,0.6604774594306946,null]},{"noun_phrase":"white car","object_id":0,"observations":[{"timestamp":4.42,"box":[0.0,0.4138888716697693,0.08802083879709244,0.5638888627290726],"occluded":false},{"timestamp":4.68,"box":[0.0,0.4342592656612396,0.23750001192092896,0.6074074059724808],"occluded":false},{"timestamp":4.94,"box":[0.11927083879709244,0.49166664481163025,0.3296875134110451,0.6481481194496155],"occluded":false},{"timestamp":5.2,"box":[0.22760418057441711,0.520370364189148,0.39895835518836975,0.6574074029922485],"occluded":false},{"timestamp":5.46,"box":[0.26979169249534607,0.5287036895751953,0.429166704416275,0.6574073880910873],"occluded":false},{"timestamp":5.72,"box":[0.29218751192092896,0.520370364189148,0.4479166865348816,0.6472222208976746],"occluded":false},{"timestamp":5.98,"box":[0.3229166865348816,0.5342592597007751,0.47708335518836975,0.6601851880550385],"occluded":false},{"timestamp":6.24,"box":[0.3604166805744171,0.5648148059844971,0.5135416835546494,0.6898148059844971],"occluded":false},{"timestamp":6.5,"box":[0.38489586114883423,0.5805555582046509,0.5354167073965073,0.7064814865589142],"occluded":false},{"timestamp":6.76,"box":[0.39635419845581055,0.5944444537162781,0.547395870089531,0.7222222238779068],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.40677085518836975,0.6092592477798462,0.5583333671092987,0.7388888746500015],"occluded":false},{"timestamp":7.28,"box":[0.4505208432674408,0.5990740656852722,0.6020833551883698,0.7324073910713196],"occluded":false},{"timestamp":7.54,"box":[0.5010417103767395,0.6157407164573669,0.6572917252779007,0.7499999701976776],"occluded":false},{"timestamp":7.8,"box":[0.5520833730697632,0.5962963104248047,0.7130208760499954,0.7370370477437973],"occluded":false},{"timestamp":8.06,"box":[0.6031250357627869,0.5703703761100769,0.7703125476837158,0.7185185253620148],"occluded":false},{"timestamp":8.32,"box":[0.6510416865348816,0.54537034034729,0.8281250298023224,0.6981481164693832],"occluded":false},{"timestamp":8.58,"box":[0.6989583969116211,0.5249999761581421,0.8979167342185974,0.6824073791503906],"occluded":false},{"timestamp":8.84,"box":[0.7734375596046448,0.4712962806224823,0.9994792342185974,0.6453703492879868],"occluded":false},{"timestamp":9.1,"box":[0.8281250596046448,0.442592591047287,0.9994792342185974,0.6231481432914734],"occluded":false},{"timestamp":9.36,"box":[0.8677083849906921,0.4462962746620178,0.9994792193174362,0.619444414973259],"occluded":false},{"timestamp":9.620000000000001,"box":[0.8734375238418579,0.4740740656852722,0.999479204416275,0.645370364189148],"occluded":false},{"timestamp":9.88,"box":[0.8234375715255737,0.4879629611968994,0.9994792491197586,0.6648148149251938],"occluded":false},{"timestamp":10.14,"box":[0.07447917014360428,0.4564814865589142,0.9354167059063911,0.682407408952713],"occluded":false}],"scores":[0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708]},{"noun_phrase":"white car","object_id":1,"observations":[{"timestamp":9.1,"box":[0.07135417312383652,0.4824073910713196,0.1015625074505806,0.5185185000300407],"occluded":false},{"timestamp":9.36,"box":[0.11510417610406876,0.47685185074806213,0.14166667684912682,0.5129629597067833],"occluded":false},{"timestamp":9.620000000000001,"box":[0.12031250447034836,0.47962963581085205,0.14739583991467953,0.5148148201406002],"occluded":false},{"timestamp":9.88,"box":[0.0677083358168602,0.47962963581085205,0.09947917237877846,0.5185185223817825],"occluded":false}],"scores":[0.8518518805503845,0.8518518805503845,0.8518518805503845,0.8518518805503845]}],"tracklets":[{"noun_phrase":"grey SUV","object_id":0,"presentation_id":1,"observations":[{"timestamp":0.0,"box":[0.9312500357627869,0.4259259104728699,0.999479204416275,0.7833333015441895],"occluded":false},{"timestamp":0.26,"box":[0.8828125596046448,0.4185185134410858,0.9994792342185974,0.787962943315506],"occluded":false},{"timestamp":0.52,"box":[0.8479167222976685,0.4175925850868225,0.9994792342185974,0.817592591047287],"occluded":false},{"timestamp":0.78,"box":[0.8796875476837158,0.4444444477558136,0.9994792193174362,0.8305555582046509],"occluded":false}],"mean_score":0.6908212304115295},{"noun_phrase":"grey SUV","object_id":1,"presentation_id":2,"observations":[{"timestamp":4.16,"box":[0.0,0.35462963581085205,0.0598958358168602,0.5009259283542633],"occluded":false},{"timestamp":4.42,"box":[0.01875000074505806,0.39814814925193787,0.23072918131947517,0.5601851791143417],"occluded":false},{"timestamp":4.68,"box":[0.2291666716337204,0.41018518805503845,0.35364584624767303,0.5453703701496124],"occluded":false},{"timestamp":4.94,"box":[0.32239586114883423,0.46759259700775146,0.43385419994592667,0.5916666686534882],"occluded":false},{"timestamp":5.2,"box":[0.3906250298023224,0.49351850152015686,0.4906250387430191,0.605555534362793],"occluded":false},{"timestamp":5.46,"box":[0.42239585518836975,0.5055555701255798,0.5161458626389503,0.6111111268401146],"occluded":false},{"timestamp":5.72,"box":[0.43541669845581055,0.4972222149372101,0.5354167073965073,0.6027777716517448],"occluded":false},{"timestamp":5.98,"box":[0.47083336114883423,0.5111110806465149,0.5645833685994148,0.6157407090067863],"occluded":false},{"timestamp":6.24,"box":[0.5067708492279053,0.5425925850868225,0.6026041880249977,0.6509259194135666],"occluded":false},{"timestamp":6.5,"box":[0.5307291746139526,0.5574073791503906,0.627604179084301,0.6648147851228714],"occluded":false},{"timestamp":6.76,"box":[0.5411458611488342,0.5675925612449646,0.6406250298023224,0.6777777448296547],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.5505208373069763,0.5814814567565918,0.651562511920929,0.6935184895992279],"occluded":false},{"timestamp":7.28,"box":[0.5921875238418579,0.5629629492759705,0.6979166939854622,0.6787036880850792],"occluded":false},{"timestamp":7.54,"box":[0.6463541984558105,0.5796296000480652,0.7567708715796471,0.6990740448236465],"occluded":false},{"timestamp":7.8,"box":[0.7010416984558105,0.550000011920929,0.8218750357627869,0.6768518686294556],"occluded":false},{"timestamp":8.06,"box":[0.7562500238418579,0.5148147940635681,0.8906250298023224,0.6499999761581421],"occluded":false},{"timestamp":8.32,"box":[0.8125000596046448,0.4861111044883728,0.965625062584877,0.6305555403232574],"occluded":false},{"timestamp":8.58,"box":[0.8791667222976685,0.4611110985279083,0.9994792267680168,0.6166666448116302],"occluded":false},{"timestamp":10.14,"box":[0.9187500476837158,0.4870370328426361,0.9994792193174362,0.6259259134531021],"occluded":false}],"mean_score":0.8439024686813354},{"noun_phrase":"white car","object_id":0,"presentation_id":3,"observations":[{"timestamp":4.42,"box":[0.0,0.4138888716697693,0.08802083879709244,0.5638888627290726],"occluded":false},{"timestamp":4.68,"box":[0.0,0.4342592656612396,0.23750001192092896,0.6074074059724808],"occluded":false},{"timestamp":4.94,"box":[0.11927083879709244,0.49166664481163025,0.3296875134110451,0.6481481194496155],"occluded":false},{"timestamp":5.2,"box":[0.22760418057441711,0.520370364189148,0.39895835518836975,0.6574074029922485],"occluded":false},{"timestamp":5.46,"box":[0.26979169249534607,0.5287036895751953,0.429166704416275,0.6574073880910873],"occluded":false},{"timestamp":5.72,"box":[0.29218751192092896,0.520370364189148,0.4479166865348816,0.6472222208976746],"occluded":false},{"timestamp":5.98,"box":[0.3229166865348816,0.5342592597007751,0.47708335518836975,0.6601851880550385],"occluded":false},{"timestamp":6.24,"box":[0.3604166805744171,0.5648148059844971,0.5135416835546494,0.6898148059844971],"occluded":false},{"timestamp":6.5,"box":[0.38489586114883423,0.5805555582046509,0.5354167073965073,0.7064814865589142],"occluded":false},{"timestamp":6.76,"box":[0.39635419845581055,0.5944444537162781,0.547395870089531,0.7222222238779068],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.40677085518836975,0.6092592477798462,0.5583333671092987,0.7388888746500015],"occluded":false},{"timestamp":7.28,"box":[0.4505208432674408,0.5990740656852722,0.6020833551883698,0.7324073910713196],"occluded":false},{"timestamp":7.54,"box":[0.5010417103767395,0.6157407164573669,0.6572917252779007,0.7499999701976776],"occluded":false},{"timestamp":7.8,"box":[0.5520833730697632,0.5962963104248047,0.7130208760499954,0.7370370477437973],"occluded":false},{"timestamp":8.06,"box":[0.6031250357627869,0.5703703761100769,0.7703125476837158,0.7185185253620148],"occluded":false},{"timestamp":8.32,"box":[0.6510416865348816,0.54537034034729,0.8281250298023224,0.6981481164693832],"occluded":false},{"timestamp":8.58,"box":[0.6989583969116211,0.5249999761581421,0.8979167342185974,0.6824073791503906],"occluded":false},{"timestamp":8.84,"box":[0.7734375596046448,0.4712962806224823,0.9994792342185974,0.6453703492879868],"occluded":false},{"timestamp":9.1,"box":[0.8281250596046448,0.442592591047287,0.9994792342185974,0.6231481432914734],"occluded":false},{"timestamp":9.36,"box":[0.8677083849906921,0.4462962746620178,0.9994792193174362,0.619444414973259],"occluded":false},{"timestamp":9.620000000000001,"box":[0.8734375238418579,0.4740740656852722,0.999479204416275,0.645370364189148],"occluded":false},{"timestamp":9.88,"box":[0.8234375715255737,0.4879629611968994,0.9994792491197586,0.6648148149251938],"occluded":false},{"timestamp":10.14,"box":[0.07447917014360428,0.4564814865589142,0.9354167059063911,0.682407408952713],"occluded":false}],"mean_score":0.941605806350708},{"noun_phrase":"white car","object_id":1,"presentation_id":4,"observations":[{"timestamp":9.1,"box":[0.07135417312383652,0.4824073910713196,0.1015625074505806,0.5185185000300407],"occluded":false},{"timestamp":9.36,"box":[0.11510417610406876,0.47685185074806213,0.14166667684912682,0.5129629597067833],"occluded":false},{"timestamp":9.620000000000001,"box":[0.12031250447034836,0.47962963581085205,0.14739583991467953,0.5148148201406002],"occluded":false},{"timestamp":9.88,"box":[0.0677083358168602,0.47962963581085205,0.09947917237877846,0.5185185223817825],"occluded":false}],"mean_score":0.8518518805503845},{"noun_phrase":"red car","object_id":0,"presentation_id":5,"observations":[{"timestamp":3.12,"box":[0.02916666865348816,0.6435185074806213,0.19427084922790527,0.8333333134651184],"occluded":false},{"timestamp":3.38,"box":[0.0390625037252903,0.6499999761581421,0.20520834997296333,0.8425925672054291],"occluded":false},{"timestamp":3.64,"box":[0.1093750074505806,0.6333333253860474,0.2645833417773247,0.8120370209217072],"occluded":false},{"timestamp":3.9,"box":[0.20208334922790527,0.6509259343147278,0.3421875238418579,0.817592591047287],"occluded":false},{"timestamp":4.16,"box":[0.35989585518836975,0.6527777910232544,0.4843750298023224,0.8064814954996109],"occluded":false},{"timestamp":4.42,"box":[0.4947916865348816,0.6527777910232544,0.622395858168602,0.8046296387910843],"occluded":false},{"timestamp":4.68,"box":[0.6088541746139526,0.6425926089286804,0.7427083551883698,0.7981481552124023],"occluded":false},{"timestamp":4.94,"box":[0.6765625476837158,0.6814814805984497,0.8208333849906921,0.8388888835906982],"occluded":false},{"timestamp":5.2,"box":[0.7192708849906921,0.6824073791503906,0.8609375655651093,0.8351851552724838],"occluded":false},{"timestamp":5.46,"box":[0.7432292103767395,0.6953703761100769,0.8906250447034836,0.8527777791023254],"occluded":false},{"timestamp":5.72,"box":[0.7645833492279053,0.6925925612449646,0.9239583611488342,0.8574073761701584],"occluded":false},{"timestamp":5.98,"box":[0.8072916865348816,0.7138888835906982,0.9885416924953461,0.8981481343507767],"occluded":false}],"mean_score":0.6604774594306946}],"dropped":{},"fates":[{"noun_phrase":"grey SUV","object_id":0,"frames":4,"mean_score":0.6908212304115295,"mean_area":0.043591340820069324,"outcome":"kept","presentation_id":1,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"grey SUV","object_id":1,"frames":19,"mean_score":0.8439024686813354,"mean_area":0.014171870340556229,"outcome":"kept","presentation_id":2,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"red car","object_id":0,"frames":12,"mean_score":0.6604774594306946,"mean_area":0.025081220095020913,"outcome":"kept","presentation_id":5,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"white car","object_id":0,"frames":23,"mean_score":0.941605806350708,"mean_area":0.0319464373135848,"outcome":"kept","presentation_id":3,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"white car","object_id":1,"frames":4,"mean_score":0.8518518805503845,"mean_area":0.00105963060503716,"outcome":"kept","presentation_id":4,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null}],"gates":{"min_track_frames":3,"min_mean_score":0.5,"merge_iou":0.8,"max_tracklets":12}}},"render":{"inputs":{"tracklets":5,"as_video":true,"fps":1.0,"max_frames":30},"outputs":{"media":"/home/zzk/data/video_bbox/vlm_gemini3.5_lite/renders/--cif7Nv6BI_chunk_000_scene_044_choices.mp4","media_bytes":3209753,"frames":[{"index":1,"timestamp":0.0,"file":"--cif7Nv6BI_chunk_000_scene_044_choices.mp4"},{"index":2,"timestamp":1.04,"file":"--cif7Nv6BI_chunk_000_scene_044_choices.mp4"},{"index":3,"timestamp":2.08,"file":"--cif7Nv6BI_chunk_000_scene_044_choices.mp4"},{"index":4,"timestamp":3.12,"file":"--cif7Nv6BI_chunk_000_scene_044_choices.mp4"},{"index":5,"timestamp":4.16,"file":"--cif7Nv6BI_chunk_000_scene_044_choices.mp4"},{"index":6,"timestamp":5.2,"file":"--cif7Nv6BI_chunk_000_scene_044_choices.mp4"},{"index":7,"timestamp":5.98,"file":"--cif7Nv6BI_chunk_000_scene_044_choices.mp4"},{"index":8,"timestamp":7.0200000000000005,"file":"--cif7Nv6BI_chunk_000_scene_044_choices.mp4"},{"index":9,"timestamp":8.06,"file":"--cif7Nv6BI_chunk_000_scene_044_choices.mp4"},{"index":10,"timestamp":9.1,"file":"--cif7Nv6BI_chunk_000_scene_044_choices.mp4"},{"index":11,"timestamp":10.14,"file":"--cif7Nv6BI_chunk_000_scene_044_choices.mp4"}]}},"select":{"inputs":{"model":"gemini-3.5-flash-lite","temperature":0.0,"as_video":true,"media_bytes":3209753,"media":["--cif7Nv6BI_chunk_000_scene_044_choices.mp4"],"choice_frames":11,"system_prompt":"You are labelling video data for referring multi-object tracking.\n\nYou will see a numbered sequence of frames from one clip, in time order. Each frame has its own number and timestamp printed in the top-left corner. Candidate objects have been detected for you and drawn on the frames as coloured boxes, each labelled with a number. The same object keeps the same number and the same colour on every frame. Below each frame is a strip of enlarged crops of those same numbered boxes, so you can identify a small or distant object.\n\nFor each description you are given, answer two questions:\n\n1. MATCHING_IDS — which of the numbered objects the description refers to. List every number that matches, or an EMPTY LIST if none of them do. An empty list is a correct and expected answer: some descriptions are deliberately about something that is not in this clip. Do not pick a number just to have picked one, and do not pick a number for an object that merely resembles the description.\n\n2. START_FRAME and END_FRAME — the frame numbers between which the described state actually holds. If the description is \"the car that's turning left\", the interval is the frames in which that car is turning, NOT every frame in which the car is visible. Use the frame numbers printed on the frames. START_FRAME must not be greater than END_FRAME. When MATCHING_IDS is empty, give the full range (the first and last frame numbers); it is ignored.\n\n3. CONFIDENCE — how sure you are of the MATCHING_IDS answer, from 0.0 to 1.0. Be honest: a low number costs the example nothing, and an overconfident wrong answer becomes a wrong label.\n\nRules:\n- Only use numbers that are actually drawn on the frames.\n- Never output coordinates, box values, or pixel positions.\n- Judge motion by comparing the object's position across frames, not by guessing from one frame.\n","user_prompt":"The video has 11 frames, numbered 1 to 11. Each frame is one moment of the clip and has its number burned into the top left corner.\n\nAnswer for each of the following descriptions, in the order given:\n1. the white car parked along the stone building\n2. the grey SUV parked next to the white car\n3. the red car parked on the street","questions":["the white car parked along the stone building","the grey SUV parked next to the white car","the red car parked on the street"]},"outputs":{"raw_response":"{\n  \"answers\": [\n    {\n      \"description_index\": 1,\n      \"matching_ids\": [\n        3\n      ],\n      \"start_frame\": 5,\n      \"end_frame\": 11,\n      \"confidence\": 1.0\n    },\n    {\n      \"description_index\": 2,\n      \"matching_ids\": [\n        2\n      ],\n      \"start_frame\": 5,\n      \"end_frame\": 11,\n      \"confidence\": 1.0\n    },\n    {\n      \"description_index\": 3,\n      \"matching_ids\": [],\n      \"start_frame\": 1,\n      \"end_frame\": 11,\n      \"confidence\": 1.0\n    }\n  ]\n}","raw_answers":[{"description_index":1,"matching_ids":[3],"start_frame":5,"end_frame":11,"confidence":1.0},{"description_index":2,"matching_ids":[2],"start_frame":5,"end_frame":11,"confidence":1.0},{"description_index":3,"matching_ids":[],"start_frame":1,"end_frame":11,"confidence":1.0}],"answers":[{"expression":"the white car parked along the stone building","matching_presentation_ids":[3],"start_choice_frame":5,"end_choice_frame":11,"confidence":1.0,"self_reported_confidence":1.0},{"expression":"the grey SUV parked next to the white car","matching_presentation_ids":[2],"start_choice_frame":5,"end_choice_frame":11,"confidence":1.0,"self_reported_confidence":1.0},{"expression":"the red car parked on the street","matching_presentation_ids":[],"start_choice_frame":1,"end_choice_frame":11,"confidence":1.0,"self_reported_confidence":1.0}]}},"verify":{"inputs":{"expressions":3,"tracklets":5,"selections":3,"gates":"interval, count, unknown ids, confidence, then geometry (motion)"},"outputs":{"kept":[{"expression":"the white car parked along the stone building","kind":"appearance","is_negative":false,"tracks":[{"timestamps":[4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88,10.14],"boxes":[[0.0,0.4138888716697693,0.08802083879709244,0.5638888627290726],[0.0,0.4342592656612396,0.23750001192092896,0.6074074059724808],[0.11927083879709244,0.49166664481163025,0.3296875134110451,0.6481481194496155],[0.22760418057441711,0.520370364189148,0.39895835518836975,0.6574074029922485],[0.26979169249534607,0.5287036895751953,0.429166704416275,0.6574073880910873],[0.29218751192092896,0.520370364189148,0.4479166865348816,0.6472222208976746],[0.3229166865348816,0.5342592597007751,0.47708335518836975,0.6601851880550385],[0.3604166805744171,0.5648148059844971,0.5135416835546494,0.6898148059844971],[0.38489586114883423,0.5805555582046509,0.5354167073965073,0.7064814865589142],[0.39635419845581055,0.5944444537162781,0.547395870089531,0.7222222238779068],[0.40677085518836975,0.6092592477798462,0.5583333671092987,0.7388888746500015],[0.4505208432674408,0.5990740656852722,0.6020833551883698,0.7324073910713196],[0.5010417103767395,0.6157407164573669,0.6572917252779007,0.7499999701976776],[0.5520833730697632,0.5962963104248047,0.7130208760499954,0.7370370477437973],[0.6031250357627869,0.5703703761100769,0.7703125476837158,0.7185185253620148],[0.6510416865348816,0.54537034034729,0.8281250298023224,0.6981481164693832],[0.6989583969116211,0.5249999761581421,0.8979167342185974,0.6824073791503906],[0.7734375596046448,0.4712962806224823,0.9994792342185974,0.6453703492879868],[0.8281250596046448,0.442592591047287,0.9994792342185974,0.6231481432914734],[0.8677083849906921,0.4462962746620178,0.9994792193174362,0.619444414973259],[0.8734375238418579,0.4740740656852722,0.999479204416275,0.645370364189148],[0.8234375715255737,0.4879629611968994,0.9994792491197586,0.6648148149251938],[0.07447917014360428,0.4564814865589142,0.9354167059063911,0.682407408952713]]}]},{"expression":"the grey SUV parked next to the white car","kind":"spatial","is_negative":false,"tracks":[{"timestamps":[4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,10.14],"boxes":[[0.0,0.35462963581085205,0.0598958358168602,0.5009259283542633],[0.01875000074505806,0.39814814925193787,0.23072918131947517,0.5601851791143417],[0.2291666716337204,0.41018518805503845,0.35364584624767303,0.5453703701496124],[0.32239586114883423,0.46759259700775146,0.43385419994592667,0.5916666686534882],[0.3906250298023224,0.49351850152015686,0.4906250387430191,0.605555534362793],[0.42239585518836975,0.5055555701255798,0.5161458626389503,0.6111111268401146],[0.43541669845581055,0.4972222149372101,0.5354167073965073,0.6027777716517448],[0.47083336114883423,0.5111110806465149,0.5645833685994148,0.6157407090067863],[0.5067708492279053,0.5425925850868225,0.6026041880249977,0.6509259194135666],[0.5307291746139526,0.5574073791503906,0.627604179084301,0.6648147851228714],[0.5411458611488342,0.5675925612449646,0.6406250298023224,0.6777777448296547],[0.5505208373069763,0.5814814567565918,0.651562511920929,0.6935184895992279],[0.5921875238418579,0.5629629492759705,0.6979166939854622,0.6787036880850792],[0.6463541984558105,0.5796296000480652,0.7567708715796471,0.6990740448236465],[0.7010416984558105,0.550000011920929,0.8218750357627869,0.6768518686294556],[0.7562500238418579,0.5148147940635681,0.8906250298023224,0.6499999761581421],[0.8125000596046448,0.4861111044883728,0.965625062584877,0.6305555403232574],[0.8791667222976685,0.4611110985279083,0.9994792267680168,0.6166666448116302],[0.9187500476837158,0.4870370328426361,0.9994792193174362,0.6259259134531021]]}]},{"expression":"the red car parked on the street","kind":"appearance","is_negative":true,"tracks":[]}],"rejected":{},"rejection_counts":{},"row":{"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_000_scene_044.mp4","scene_key":"--cif7Nv6BI_chunk_000_scene_044","category":"test_vehicles","height":1080,"width":1920,"fps":50.0,"num_frames":516,"duration":10.32,"rmot_expressions":[{"expression":"the white car parked along the stone building","kind":"appearance","is_negative":false,"tracks":[{"timestamps":[4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88,10.14],"boxes":[[0.0,0.4138888716697693,0.08802083879709244,0.5638888627290726],[0.0,0.4342592656612396,0.23750001192092896,0.6074074059724808],[0.11927083879709244,0.49166664481163025,0.3296875134110451,0.6481481194496155],[0.22760418057441711,0.520370364189148,0.39895835518836975,0.6574074029922485],[0.26979169249534607,0.5287036895751953,0.429166704416275,0.6574073880910873],[0.29218751192092896,0.520370364189148,0.4479166865348816,0.6472222208976746],[0.3229166865348816,0.5342592597007751,0.47708335518836975,0.6601851880550385],[0.3604166805744171,0.5648148059844971,0.5135416835546494,0.6898148059844971],[0.38489586114883423,0.5805555582046509,0.5354167073965073,0.7064814865589142],[0.39635419845581055,0.5944444537162781,0.547395870089531,0.7222222238779068],[0.40677085518836975,0.6092592477798462,0.5583333671092987,0.7388888746500015],[0.4505208432674408,0.5990740656852722,0.6020833551883698,0.7324073910713196],[0.5010417103767395,0.6157407164573669,0.6572917252779007,0.7499999701976776],[0.5520833730697632,0.5962963104248047,0.7130208760499954,0.7370370477437973],[0.6031250357627869,0.5703703761100769,0.7703125476837158,0.7185185253620148],[0.6510416865348816,0.54537034034729,0.8281250298023224,0.6981481164693832],[0.6989583969116211,0.5249999761581421,0.8979167342185974,0.6824073791503906],[0.7734375596046448,0.4712962806224823,0.9994792342185974,0.6453703492879868],[0.8281250596046448,0.442592591047287,0.9994792342185974,0.6231481432914734],[0.8677083849906921,0.4462962746620178,0.9994792193174362,0.619444414973259],[0.8734375238418579,0.4740740656852722,0.999479204416275,0.645370364189148],[0.8234375715255737,0.4879629611968994,0.9994792491197586,0.6648148149251938],[0.07447917014360428,0.4564814865589142,0.9354167059063911,0.682407408952713]]}]},{"expression":"the grey SUV parked next to the white car","kind":"spatial","is_negative":false,"tracks":[{"timestamps":[4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,10.14],"boxes":[[0.0,0.35462963581085205,0.0598958358168602,0.5009259283542633],[0.01875000074505806,0.39814814925193787,0.23072918131947517,0.5601851791143417],[0.2291666716337204,0.41018518805503845,0.35364584624767303,0.5453703701496124],[0.32239586114883423,0.46759259700775146,0.43385419994592667,0.5916666686534882],[0.3906250298023224,0.49351850152015686,0.4906250387430191,0.605555534362793],[0.42239585518836975,0.5055555701255798,0.5161458626389503,0.6111111268401146],[0.43541669845581055,0.4972222149372101,0.5354167073965073,0.6027777716517448],[0.47083336114883423,0.5111110806465149,0.5645833685994148,0.6157407090067863],[0.5067708492279053,0.5425925850868225,0.6026041880249977,0.6509259194135666],[0.5307291746139526,0.5574073791503906,0.627604179084301,0.6648147851228714],[0.5411458611488342,0.5675925612449646,0.6406250298023224,0.6777777448296547],[0.5505208373069763,0.5814814567565918,0.651562511920929,0.6935184895992279],[0.5921875238418579,0.5629629492759705,0.6979166939854622,0.6787036880850792],[0.6463541984558105,0.5796296000480652,0.7567708715796471,0.6990740448236465],[0.7010416984558105,0.550000011920929,0.8218750357627869,0.6768518686294556],[0.7562500238418579,0.5148147940635681,0.8906250298023224,0.6499999761581421],[0.8125000596046448,0.4861111044883728,0.965625062584877,0.6305555403232574],[0.8791667222976685,0.4611110985279083,0.9994792267680168,0.6166666448116302],[0.9187500476837158,0.4870370328426361,0.9994792193174362,0.6259259134531021]]}]},{"expression":"the red car parked on the street","kind":"appearance","is_negative":true,"tracks":[]}],"rmot_expression_count":3,"rmot_negative_count":1,"rmot_motion_count":0}}}}},{"index":3,"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_014_scene_002.mp4","status":"complete","steps":{"decode":{"inputs":{"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_014_scene_002.mp4","sample_fps":4.0},"outputs":{"width":1920,"height":1080,"source_fps":50.0,"duration_seconds":15.4,"frames_kept":60,"blank_frames_dropped":0,"timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88,10.14,10.4,10.66,10.92,11.18,11.44,11.700000000000001,11.96,12.22,12.48,12.74,13.0,13.26,13.52,13.780000000000001,14.040000000000001,14.3,14.56,14.82,15.08,15.34],"error":""}},"propose":{"inputs":{"model":"gemini-3.5-flash-lite","temperature":0.4,"video_bytes":18605944,"system_prompt":"You are annotating video clips to build training data for referring multi-object tracking: given a video and a natural-language description, a model must output the boxes of every object matching that description, over time.\n\nYour job is to write the descriptions. You do NOT output coordinates, boxes, pixel positions, or any numbers about where things are. A separate detector finds the objects.\n\nFor every clip you produce a list of items. Each item has two texts that must NOT be the same:\n\nEXPRESSION — what a person would say to single out the object or objects. It may describe motion, temporal order, interaction with other objects, and spatial relations. It should read like something a person would actually say.\n  Good: \"the black car that's turning left\"\n  Good: \"the cyclist who overtakes the bus\"\n  Good: \"the two people who walk out of frame on the right\"\n\nNOUN_PHRASE — the same object as a plain, timeless visual category with its visible attributes, and NOTHING else. The detector is prompted with this text and looks at ONE frame at a time, so anything about time, order, or action is useless to it and makes it fail.\n  Allowed: category nouns, colours, materials, clothing, size, type\n  Forbidden: verbs and participles of action (\"turning\", \"walks\", \"is moving\"), temporal words (\"then\", \"starts\", \"after\", \"first\", \"last\"), relative clauses (\"that ...\", \"which ...\", \"who ...\"), ordinals and rank words (\"first\", \"leftmost\", \"nearest\"), and anything over six words.\n  \"the black car that's turning left\" -> \"black car\"\n  \"the cyclist who overtakes the bus\" -> \"cyclist\"\n  \"the two people who walk out of frame on the right\" -> \"person\"\n\nKIND — what makes the expression hard: \"motion\", \"appearance\", \"spatial\", or \"interaction\". Aim for about 35% \"motion\" items: motion is the part a single-frame model cannot already solve, so it is the most valuable, but a list of only motion items is not usable training data either.\n\nEXPECTED_COUNT — how many distinct objects in the clip match the EXPRESSION. Count carefully; this is checked against the detector's answer, and a wrong count discards the item.\n\nIS_NEGATIVE — at most ONE item may be a negative, and only when you also write at least one non-negative item. A negative is an expression that recombines attributes that ARE present in the clip into a whole that is NOT. If the only black car is moving, \"the parked black car\" is a good negative. Its NOUN_PHRASE must still be the category that IS present (\"black car\"), because that is what makes the absence checkable rather than merely unverified. A negative's EXPECTED_COUNT is 0.\n\nRules that discard an item, so follow them exactly:\n- Never write coordinates, box values, pixel positions, or percentages of the frame.\n- Never let a NOUN_PHRASE contain a verb of action, a temporal word, a relative clause, or an ordinal.\n- Never write an EXPRESSION whose object is not visible in the clip, unless it is marked IS_NEGATIVE.\n- Prefer objects that are clearly visible and identifiable over background clutter.\n- Do not describe the camera or the shot; describe objects in the scene.\n","user_prompt":"Watch this clip and write between 1 and 3 items following the rules. Cover different objects rather than restating one object several ways, and include any negative in the same list.\n"},"outputs":{"raw_response":"{\n  \"items\": [\n    {\n      \"expression\": \"the stone wall illuminated by the light\",\n      \"noun_phrase\": \"stone wall\",\n      \"kind\": \"appearance\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the wooden beams of the ruined ceiling\",\n      \"noun_phrase\": \"wooden beam\",\n      \"kind\": \"appearance\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the modern metal railing\",\n      \"noun_phrase\": \"metal railing\",\n      \"kind\": \"appearance\",\n      \"expected_count\": 0,\n      \"is_negative\": true\n    }\n  ]\n}","raw_items":[{"expression":"the stone wall illuminated by the light","noun_phrase":"stone wall","kind":"appearance","expected_count":1,"is_negative":false},{"expression":"the wooden beams of the ruined ceiling","noun_phrase":"wooden beam","kind":"appearance","expected_count":1,"is_negative":false},{"expression":"the modern metal railing","noun_phrase":"metal railing","kind":"appearance","expected_count":0,"is_negative":true}],"accepted":[{"expression":"the stone wall illuminated by the light","noun_phrase":"stone wall","kind":"appearance","expected_count":1,"is_negative":false},{"expression":"the wooden beams of the ruined ceiling","noun_phrase":"wooden beam","kind":"appearance","expected_count":1,"is_negative":false},{"expression":"the modern metal railing","noun_phrase":"metal railing","kind":"appearance","expected_count":0,"is_negative":true}],"gated":[]}},"track":{"inputs":{"tracker":"Sam31MultiplexTracker","model_id":"/mnt/ceph3/ec/jacky/sam3.1","noun_phrases":["metal railing","stone wall","wooden beam"],"expressions_in":3,"frames":60,"sample_fps":4.0,"frame_size":"1920x1080","timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88,10.14,10.4,10.66,10.92,11.18,11.44,11.700000000000001,11.96,12.22,12.48,12.74,13.0,13.26,13.52,13.780000000000001,14.040000000000001,14.3,14.56,14.82,15.08,15.34],"session":"one session over 60 JPEG frames; then per phrase: reset_session, add_prompt(text) on frame 0, propagate_in_video","passes":3},"outputs":{"raw_tracks":[{"noun_phrase":"stone wall","object_id":0,"observations":[{"timestamp":0.0,"box":[0.6265625357627869,0.032407406717538834,0.9994792342185974,0.9694443978369236],"occluded":false},{"timestamp":0.26,"box":[0.6765625476837158,0.031481482088565826,0.9994792342185974,0.9703703299164772],"occluded":false},{"timestamp":0.52,"box":[0.6458333730697632,0.03055555559694767,0.9994792342185974,0.9712962601333857],"occluded":false},{"timestamp":0.78,"box":[0.6057292222976685,0.03055555559694767,0.9994792342185974,0.9712962601333857],"occluded":false},{"timestamp":1.04,"box":[0.6156250238418579,0.031481482088565826,0.999479204416275,0.9712962880730629],"occluded":false},{"timestamp":1.3,"box":[0.6270833611488342,0.03055555559694767,0.999479204416275,0.9712962601333857],"occluded":false},{"timestamp":1.56,"box":[0.635937511920929,0.03055555559694767,0.999479204416275,0.9712962601333857],"occluded":false},{"timestamp":1.82,"box":[0.004687500186264515,0.025925925001502037,0.9994792463257909,0.9712963011115789],"occluded":false},{"timestamp":2.08,"box":[0.47708335518836975,0.03055555559694767,0.999479204416275,0.9712962601333857],"occluded":false},{"timestamp":2.34,"box":[0.5541666746139526,0.031481482088565826,0.999479204416275,0.9712962880730629],"occluded":false},{"timestamp":2.6,"box":[0.6270833611488342,0.031481482088565826,0.999479204416275,0.9712962880730629],"occluded":false},{"timestamp":2.86,"box":[0.5328125357627869,0.031481482088565826,0.9994792342185974,0.9722221866250038],"occluded":false},{"timestamp":3.12,"box":[0.40781253576278687,0.031481482088565826,0.9994792342185974,0.9712962880730629],"occluded":false},{"timestamp":3.38,"box":[0.6776041984558105,0.031481482088565826,0.999479204416275,0.9712962880730629],"occluded":false},{"timestamp":3.64,"box":[0.26875001192092896,0.031481482088565826,0.9994792342185974,0.9712962880730629],"occluded":false},{"timestamp":3.9,"box":[0.3661458492279053,0.031481482088565826,0.9994792342185974,0.9703703299164772],"occluded":false},{"timestamp":4.16,"box":[0.6338542103767395,0.03055555559694767,0.9994792342185974,0.9712962601333857],"occluded":false},{"timestamp":4.42,"box":[0.5973958373069763,0.03055555559694767,0.999479204416275,0.9712962601333857],"occluded":false},{"timestamp":4.68,"box":[0.6432291865348816,0.031481482088565826,0.999479204416275,0.9703703299164772],"occluded":false},{"timestamp":4.94,"box":[0.643750011920929,0.03055555559694767,0.999479204416275,0.9712962601333857],"occluded":false},{"timestamp":5.2,"box":[0.6541666984558105,0.03055555559694767,0.999479204416275,0.9712962601333857],"occluded":false},{"timestamp":5.46,"box":[0.5604166984558105,0.03055555559694767,0.9994792342185974,0.9712962601333857],"occluded":false},{"timestamp":5.72,"box":[0.5385416746139526,0.03055555559694767,0.999479204416275,0.9712962601333857],"occluded":false},{"timestamp":5.98,"box":[0.6588541865348816,0.03055555559694767,0.999479204416275,0.9712962601333857],"occluded":false},{"timestamp":6.24,"box":[0.6703125238418579,0.03055555559694767,0.999479204416275,0.9712962601333857],"occluded":false},{"timestamp":6.5,"box":[0.6979166865348816,0.03055555559694767,0.999479204416275,0.9712962601333857],"occluded":false},{"timestamp":6.76,"box":[0.8104166984558105,0.03055555559694767,0.999479204416275,0.9712962601333857],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.8984375596046448,0.03055555559694767,0.9994792342185974,0.9712962601333857],"occluded":false}],"scores":[0.7928802371025085,0.7928802371025085,0.7928802371025085,0.7928802371025085,0.7928802371025085,0.7928802371025085,0.7928802371025085,0.7928802371025085,0.7928802371025085,0.7928802371025085,0.7928802371025085,0.7928802371025085,0.7928802371025085,0.7928802371025085,0.7928802371025085,0.7928802371025085,0.7928802371025085,0.7928802371025085,0.7928802371025085,0.7928802371025085,0.7928802371025085,0.7928802371025085,0.7928802371025085,0.7928802371025085,0.7928802371025085,0.7928802371025085,0.7928802371025085,0.7928802371025085]},{"noun_phrase":"stone wall","object_id":1,"observations":[{"timestamp":0.0,"box":[0.0,0.031481482088565826,0.21145834028720856,0.21388889104127884],"occluded":false},{"timestamp":0.26,"box":[0.06145833805203438,0.03055555559694767,0.2604166753590107,0.23425925709307194],"occluded":false},{"timestamp":0.52,"box":[0.0,0.03055555559694767,0.19739584624767303,0.1999999973922968],"occluded":false},{"timestamp":0.78,"box":[0.0,0.03055555559694767,0.12968750298023224,0.15277777798473835],"occluded":false},{"timestamp":1.04,"box":[0.0,0.03055555559694767,0.11666667461395264,0.1305555570870638],"occluded":false},{"timestamp":1.3,"box":[0.0,0.03055555559694767,0.30000001192092896,0.17962963320314884],"occluded":false},{"timestamp":1.56,"box":[0.0,0.031481482088565826,0.0078125,0.0842592604458332],"occluded":false}],"scores":[0.6878048777580261,0.6878048777580261,0.6878048777580261,0.6878048777580261,0.6878048777580261,0.6878048777580261,0.6878048777580261]},{"noun_phrase":"stone wall","object_id":2,"observations":[{"timestamp":2.08,"box":[0.3031250238418579,0.10925925523042679,0.5052083730697632,0.9703703597187996],"occluded":false},{"timestamp":2.34,"box":[0.30156251788139343,0.03703703731298447,0.528645858168602,0.9712962731719017],"occluded":false},{"timestamp":2.6,"box":[0.18541668355464935,0.031481482088565826,0.5604167133569717,0.9712962880730629],"occluded":false},{"timestamp":2.86,"box":[0.18541668355464935,0.031481482088565826,0.5614583641290665,0.9712962880730629],"occluded":false},{"timestamp":3.12,"box":[0.11614584177732468,0.03055555559694767,0.8244792148470879,0.9712962601333857],"occluded":false},{"timestamp":3.38,"box":[0.08958333730697632,0.03055555559694767,0.44479167461395264,0.9712962601333857],"occluded":false},{"timestamp":3.64,"box":[0.00937500037252903,0.031481482088565826,0.4197916928678751,0.9712962880730629],"occluded":false},{"timestamp":3.9,"box":[0.0,0.031481482088565826,0.3994791805744171,0.9712962880730629],"occluded":false},{"timestamp":4.16,"box":[0.0,0.03055555559694767,0.3307291865348816,0.969444403424859],"occluded":false},{"timestamp":4.42,"box":[0.0,0.03055555559694767,0.27760419249534607,0.9703703615814447],"occluded":false},{"timestamp":4.68,"box":[0.0,0.031481482088565826,0.22500000894069672,0.9712962880730629],"occluded":false},{"timestamp":4.94,"box":[0.0,0.031481482088565826,0.22604167461395264,0.9990740790963173],"occluded":false},{"timestamp":5.2,"box":[0.0,0.03055555559694767,0.23593750596046448,0.9712962601333857],"occluded":false},{"timestamp":5.46,"box":[0.0,0.03055555559694767,0.22552084922790527,0.9712962601333857],"occluded":false},{"timestamp":5.72,"box":[0.0,0.03055555559694767,0.22447918355464935,0.9712962601333857],"occluded":false},{"timestamp":5.98,"box":[0.0,0.03055555559694767,0.23125001788139343,0.9712962601333857],"occluded":false},{"timestamp":6.24,"box":[0.0,0.03055555559694767,0.2291666716337204,0.9712962601333857],"occluded":false},{"timestamp":6.5,"box":[0.0,0.03055555559694767,0.23281250894069672,0.9712962601333857],"occluded":false},{"timestamp":6.76,"box":[0.0,0.029629629105329514,0.2786458432674408,0.9712962917983532],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.0,0.03055555559694767,0.28072917461395264,0.9712962601333857],"occluded":false},{"timestamp":7.28,"box":[0.0,0.03055555559694767,0.2593750059604645,0.9712962601333857],"occluded":false},{"timestamp":7.54,"box":[0.0,0.2620370388031006,0.9078125357627869,0.9712963104248047],"occluded":false},{"timestamp":7.8,"box":[0.0,0.4037036895751953,0.8104166984558105,0.9722222089767456],"occluded":false},{"timestamp":8.06,"box":[0.0,0.5157407522201538,0.9494792222976685,0.9712963104248047],"occluded":false},{"timestamp":8.32,"box":[0.0,0.48055553436279297,0.2239583432674408,0.9712962806224823],"occluded":false},{"timestamp":8.58,"box":[0.0,0.5388888716697693,0.21822917461395264,0.9712962806224823],"occluded":false},{"timestamp":8.84,"box":[0.0,0.6037036776542664,0.20937500894069672,0.9712962508201599],"occluded":false},{"timestamp":9.1,"box":[0.0,0.6638888716697693,0.1979166716337204,0.9712962806224823],"occluded":false},{"timestamp":9.36,"box":[0.0,0.7064814567565918,0.19062501192092896,0.9712962806224823],"occluded":false},{"timestamp":9.620000000000001,"box":[0.0,0.08148147910833359,0.18593750894069672,0.9712962731719017],"occluded":false},{"timestamp":9.88,"box":[0.0,0.7074074149131775,0.05312500149011612,0.972222238779068],"occluded":false},{"timestamp":10.14,"box":[0.0,0.7694444060325623,0.05781250447034836,0.9722221791744232],"occluded":false},{"timestamp":10.66,"box":[0.0,0.47962963581085205,0.03020833432674408,0.9703703820705414],"occluded":false},{"timestamp":10.92,"box":[0.0,0.032407406717538834,0.07083334028720856,0.9722222127020359],"occluded":false},{"timestamp":11.18,"box":[0.0,0.03055555559694767,0.07916667312383652,0.9518518541008234],"occluded":false},{"timestamp":11.44,"box":[0.0,0.03055555559694767,0.0859375074505806,0.9722222182899714],"occluded":false},{"timestamp":11.700000000000001,"box":[0.0,0.03055555559694767,0.12447917461395264,0.9722222182899714],"occluded":false},{"timestamp":11.96,"box":[0.0,0.03055555559694767,0.15156251192092896,0.9722222182899714],"occluded":false},{"timestamp":12.22,"box":[0.0,0.03055555559694767,0.17395834624767303,0.9722222182899714],"occluded":false},{"timestamp":12.48,"box":[0.0,0.031481482088565826,0.17343750596046448,0.9722221866250038],"occluded":false},{"timestamp":12.74,"box":[0.0,0.03055555559694767,0.16354167461395264,0.9712962601333857],"occluded":false},{"timestamp":13.0,"box":[0.0,0.03055555559694767,0.15572917461395264,0.9712962601333857],"occluded":false},{"timestamp":13.26,"box":[0.0,0.03055555559694767,0.15989583730697632,0.9712962601333857],"occluded":false},{"timestamp":13.52,"box":[0.0,0.03055555559694767,0.16875000298023224,0.9712962601333857],"occluded":false},{"timestamp":13.780000000000001,"box":[0.0,0.03055555559694767,0.17135417461395264,0.9712962601333857],"occluded":false},{"timestamp":14.040000000000001,"box":[0.0,0.03055555559694767,0.17135417461395264,0.9712962601333857],"occluded":false},{"timestamp":14.3,"box":[0.0,0.03055555559694767,0.1979166716337204,0.9712962601333857],"occluded":false},{"timestamp":14.56,"box":[0.0,0.03055555559694767,0.19166667759418488,0.9712962601333857],"occluded":false},{"timestamp":14.82,"box":[0.0,0.03055555559694767,0.15468750894069672,0.7685185167938471],"occluded":false},{"timestamp":15.08,"box":[0.0,0.029629629105329514,0.10364583879709244,0.9703703336417675],"occluded":false}],"scores":[0.6640419960021973,0.6640419960021973,0.6640419960021973,0.6640419960021973,0.6640419960021973,0.6640419960021973,0.6640419960021973,0.6640419960021973,0.6640419960021973,0.6640419960021973,0.6640419960021973,0.6640419960021973,0.6640419960021973,0.6640419960021973,0.6640419960021973,0.6640419960021973,0.6640419960021973,0.6640419960021973,0.6640419960021973,0.6640419960021973,0.6640419960021973,0.6640419960021973,0.6640419960021973,0.6640419960021973,0.6640419960021973,0.6640419960021973,0.6640419960021973,0.6640419960021973,0.6640419960021973,0.6640419960021973,0.6640419960021973,0.6640419960021973,0.6640419960021973,0.6640419960021973,0.6640419960021973,0.6640419960021973,0.6640419960021973,0.6640419960021973,0.6640419960021973,0.6640419960021973,0.6640419960021973,0.6640419960021973,0.6640419960021973,0.6640419960021973,0.6640419960021973,0.6640419960021973,0.6640419960021973,0.6640419960021973,0.6640419960021973,0.6640419960021973]},{"noun_phrase":"wooden beam","object_id":0,"observations":[{"timestamp":5.72,"box":[0.2031250149011612,0.575925886631012,0.5057291835546494,0.6083332933485508],"occluded":false},{"timestamp":5.98,"box":[0.21197918057441711,0.5805555582046509,0.5125000178813934,0.6111111138015985],"occluded":false},{"timestamp":6.24,"box":[0.20625001192092896,0.5972222089767456,0.5140625238418579,0.6268518380820751],"occluded":false},{"timestamp":6.5,"box":[0.2057291716337204,0.5907407402992249,0.5244791954755783,0.6222222223877907],"occluded":false},{"timestamp":6.76,"box":[0.2421875149011612,0.5861111283302307,0.583333358168602,0.6148148309439421],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.21250000596046448,0.5018518567085266,0.6119791865348816,0.5490740798413754],"occluded":false},{"timestamp":7.28,"box":[0.28125,0.44999998807907104,0.6255208551883698,0.4870370253920555],"occluded":false},{"timestamp":7.54,"box":[0.2515625059604645,0.3611111044883728,0.7828125059604645,0.5740740597248077],"occluded":false},{"timestamp":7.8,"box":[0.23229168355464935,0.3305555582046509,0.7828125208616257,0.5879629552364349],"occluded":false},{"timestamp":8.06,"box":[0.22812500596046448,0.2620370388031006,0.6307291984558105,0.31203703954815865],"occluded":false},{"timestamp":8.32,"box":[0.25052085518836975,0.2342592477798462,0.6546875536441803,0.2805555425584316],"occluded":false},{"timestamp":8.58,"box":[0.24114584922790527,0.22499999403953552,0.6604166924953461,0.26759258657693863],"occluded":false},{"timestamp":8.84,"box":[0.24427084624767303,0.23148147761821747,0.6604167073965073,0.27222221717238426],"occluded":false},{"timestamp":9.1,"box":[0.24062500894069672,0.2481481432914734,0.6468750387430191,0.2935185134410858],"occluded":false},{"timestamp":9.36,"box":[0.23229168355464935,0.2759259343147278,0.6442708820104599,0.32407408207654953],"occluded":false},{"timestamp":9.620000000000001,"box":[0.24635417759418488,0.3444444239139557,0.6562500447034836,0.3990740515291691],"occluded":false},{"timestamp":9.88,"box":[0.29479166865348816,0.46296295523643494,0.6994791924953461,0.7546296119689941],"occluded":false},{"timestamp":10.14,"box":[0.30937501788139343,0.5740740895271301,0.6859375238418579,0.6259259395301342],"occluded":false},{"timestamp":10.4,"box":[0.0052083334885537624,0.6962962746620178,0.963541706558317,0.9990740418434143],"occluded":false},{"timestamp":10.66,"box":[0.06041666865348816,0.7685185074806213,0.9494791924953461,0.9990740567445755],"occluded":false},{"timestamp":10.92,"box":[0.06041666865348816,0.032407406717538834,0.8104167282581329,0.9990740455687046],"occluded":false},{"timestamp":11.18,"box":[0.03281250223517418,0.8305555582046509,0.8802083991467953,0.9990740716457367],"occluded":false},{"timestamp":11.44,"box":[0.29635417461395264,0.8638888597488403,0.9078125357627869,0.9990740418434143],"occluded":false},{"timestamp":11.700000000000001,"box":[0.26875001192092896,0.8925925493240356,0.7677083611488342,0.9990740269422531],"occluded":false},{"timestamp":11.96,"box":[0.3369791805744171,0.9351851940155029,0.7723958790302277,0.9722222313284874],"occluded":false}],"scores":[0.6683937907218933,0.6683937907218933,0.6683937907218933,0.6683937907218933,0.6683937907218933,0.6683937907218933,0.6683937907218933,0.6683937907218933,0.6683937907218933,0.6683937907218933,0.6683937907218933,0.6683937907218933,0.6683937907218933,0.6683937907218933,0.6683937907218933,0.6683937907218933,0.6683937907218933,0.6683937907218933,0.6683937907218933,0.6683937907218933,0.6683937907218933,0.6683937907218933,0.6683937907218933,0.6683937907218933,0.6683937907218933]},{"noun_phrase":"wooden beam","object_id":1,"observations":[{"timestamp":5.72,"box":[0.5135416984558105,0.03055555559694767,0.5755208693444729,0.6305555198341608],"occluded":false},{"timestamp":5.98,"box":[0.5197917222976685,0.03055555559694767,0.5786458924412727,0.6407407019287348],"occluded":false},{"timestamp":6.24,"box":[0.5223958492279053,0.03055555559694767,0.5828125178813934,0.6564814541488886],"occluded":false},{"timestamp":6.5,"box":[0.5354167222976685,0.03055555559694767,0.5953125581145287,0.6518518421798944],"occluded":false},{"timestamp":6.76,"box":[0.5953125357627869,0.03055555559694767,0.6453125402331352,0.6435185167938471],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.6244791746139526,0.03055555559694767,0.6755208447575569,0.5601852033287287],"occluded":false},{"timestamp":7.28,"box":[0.6395833492279053,0.031481482088565826,0.6958333514630795,0.5120370164513588],"occluded":false},{"timestamp":7.54,"box":[0.6291667222976685,0.031481482088565826,0.7828125655651093,0.5740740671753883],"occluded":false},{"timestamp":7.8,"box":[0.6338542103767395,0.031481482088565826,0.7968750596046448,0.5879629626870155],"occluded":false},{"timestamp":8.06,"box":[0.643750011920929,0.03055555559694767,0.6958333477377892,0.5324074123054743],"occluded":false},{"timestamp":8.32,"box":[0.31041666865348816,0.03055555559694767,0.721875011920929,0.6574074123054743],"occluded":false},{"timestamp":8.58,"box":[0.2828125059604645,0.03055555559694767,0.7312500178813934,0.6574074123054743],"occluded":false},{"timestamp":8.84,"box":[0.29635417461395264,0.031481482088565826,0.7328125238418579,0.6574073806405067],"occluded":false},{"timestamp":9.1,"box":[0.26875001192092896,0.03055555559694767,0.7187500298023224,0.6990740392357111],"occluded":false},{"timestamp":9.36,"box":[0.31041666865348816,0.031481482088565826,0.7177083492279053,0.7268518581986427],"occluded":false},{"timestamp":9.620000000000001,"box":[0.29635417461395264,0.03055555559694767,0.7322916984558105,0.7962963078171015],"occluded":false},{"timestamp":9.88,"box":[0.32447919249534607,0.031481482088565826,0.7588542103767395,0.9351851716637611],"occluded":false},{"timestamp":10.14,"box":[0.7109375596046448,0.031481482088565826,0.7567708939313889,0.6898147836327553],"occluded":false},{"timestamp":10.4,"box":[0.01875000074505806,0.029629629105329514,0.9635417349636555,0.9990740828216076],"occluded":false},{"timestamp":10.66,"box":[0.06041666865348816,0.03055555559694767,0.9911458790302277,0.99907405115664],"occluded":false},{"timestamp":10.92,"box":[0.03281250223517418,0.03055555559694767,0.9911458753049374,0.99907405115664],"occluded":false},{"timestamp":11.18,"box":[0.03281250223517418,0.031481482088565826,0.9494791887700558,0.9990740790963173],"occluded":false},{"timestamp":11.44,"box":[0.07447917014360428,0.032407406717538834,0.9911458566784859,0.9990740455687046],"occluded":false},{"timestamp":11.700000000000001,"box":[0.33802086114883423,0.031481482088565826,0.99114590883255,0.9990740790963173],"occluded":false},{"timestamp":11.96,"box":[0.7385417222976685,0.043518517166376114,0.8270833939313889,0.9712962843477726],"occluded":false},{"timestamp":12.22,"box":[0.3802083432674408,0.11759258806705475,0.8296875357627869,0.9842592626810074],"occluded":false},{"timestamp":12.48,"box":[0.5604166984558105,0.21851851046085358,0.8223958909511566,0.9990740567445755],"occluded":false},{"timestamp":12.74,"box":[0.2828125059604645,0.3092592656612396,0.807291716337204,0.9990740716457367],"occluded":false},{"timestamp":13.0,"box":[0.15781250596046448,0.37870368361473083,0.7947917282581329,0.9990740716457367],"occluded":false},{"timestamp":13.26,"box":[0.24114584922790527,0.4342592656612396,0.7843750715255737,0.9990740716457367],"occluded":false},{"timestamp":13.52,"box":[0.4489583671092987,0.4824073910713196,0.776041716337204,0.9990740418434143],"occluded":false},{"timestamp":13.780000000000001,"box":[0.3661458492279053,0.5074074268341064,0.7614583671092987,0.9990740716457367],"occluded":false},{"timestamp":14.040000000000001,"box":[0.26875001192092896,0.5166666507720947,0.7536458671092987,0.9990740418434143],"occluded":false},{"timestamp":14.3,"box":[0.6822916865348816,0.5324074029922485,0.747395858168602,0.970370352268219],"occluded":false},{"timestamp":14.56,"box":[0.6562500596046448,0.5462962985038757,0.7104167304933071,0.970370352268219],"occluded":false},{"timestamp":14.82,"box":[0.6020833849906921,0.5324074029922485,0.6421875543892384,0.9712962806224823],"occluded":false},{"timestamp":15.08,"box":[0.15781250596046448,0.046296294778585434,0.5687500238418579,0.9703703485429287],"occluded":false},{"timestamp":15.34,"box":[0.32447919249534607,0.4435185194015503,0.40781252831220627,0.9712963104248047],"occluded":false}],"scores":[0.6995305418968201,0.6995305418968201,0.6995305418968201,0.6995305418968201,0.6995305418968201,0.6995305418968201,0.6995305418968201,0.6995305418968201,0.6995305418968201,0.6995305418968201,0.6995305418968201,0.6995305418968201,0.6995305418968201,0.6995305418968201,0.6995305418968201,0.6995305418968201,0.6995305418968201,0.6995305418968201,0.6995305418968201,0.6995305418968201,0.6995305418968201,0.6995305418968201,0.6995305418968201,0.6995305418968201,0.6995305418968201,0.6995305418968201,0.6995305418968201,0.6995305418968201,0.6995305418968201,0.6995305418968201,0.6995305418968201,0.6995305418968201,0.6995305418968201,0.6995305418968201,0.6995305418968201,0.6995305418968201,0.6995305418968201,0.6995305418968201]},{"noun_phrase":"wooden beam","object_id":2,"observations":[{"timestamp":7.28,"box":[0.32604169845581055,0.6370370388031006,0.8020834028720856,0.7212962955236435],"occluded":false},{"timestamp":7.54,"box":[0.31041666865348816,0.5537036657333374,0.788541704416275,0.6425925493240356],"occluded":false},{"timestamp":7.8,"box":[0.2994791865348816,0.5314814448356628,0.8109375238418579,0.623148113489151],"occluded":false},{"timestamp":8.06,"box":[0.2291666716337204,0.46666666865348816,0.8250000327825546,0.565740741789341],"occluded":false},{"timestamp":8.32,"box":[0.25468751788139343,0.4546296298503876,0.858854204416275,0.6574074029922485],"occluded":false},{"timestamp":8.58,"box":[0.2828125059604645,0.45092591643333435,0.8718750178813934,0.6574073880910873],"occluded":false},{"timestamp":8.84,"box":[0.29635417461395264,0.4592592418193817,0.8723958730697632,0.6712962687015533],"occluded":false},{"timestamp":9.1,"box":[0.26875001192092896,0.4740740656852722,0.8604167103767395,0.6574074029922485],"occluded":false},{"timestamp":9.36,"box":[0.31197917461395264,0.4962962865829468,0.8572916984558105,0.5935185104608536],"occluded":false},{"timestamp":9.620000000000001,"box":[0.3369791805744171,0.5722222328186035,0.8750000298023224,0.6722222343087196],"occluded":false},{"timestamp":9.88,"box":[0.34322917461395264,0.7037037014961243,0.9041666984558105,0.8074074015021324],"occluded":false},{"timestamp":10.14,"box":[0.3671875298023224,0.8481481075286865,0.9218750298023224,0.9490740299224854],"occluded":false}],"scores":[0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563]},{"noun_phrase":"wooden beam","object_id":3,"observations":[{"timestamp":7.8,"box":[0.18281251192092896,0.6435185074806213,0.30625002086162567,0.8009259104728699],"occluded":false},{"timestamp":8.06,"box":[0.1875000149011612,0.585185170173645,0.30781251937150955,0.7388888746500015],"occluded":false},{"timestamp":8.32,"box":[0.2031250149011612,0.5435184836387634,0.35572919249534607,0.7074073702096939],"occluded":false},{"timestamp":8.58,"box":[0.20520834624767303,0.5324074029922485,0.32760418951511383,0.6990740597248077],"occluded":false},{"timestamp":8.84,"box":[0.20260417461395264,0.550000011920929,0.3244791775941849,0.7074074149131775],"occluded":false},{"timestamp":9.1,"box":[0.19114583730697632,0.565740704536438,0.3166666775941849,0.7296295911073685],"occluded":false},{"timestamp":9.36,"box":[0.18906250596046448,0.5972222089767456,0.31562501192092896,0.7564814686775208],"occluded":false},{"timestamp":9.620000000000001,"box":[0.2083333432674408,0.6777777671813965,0.3291666805744171,0.8333333134651184],"occluded":false},{"timestamp":9.88,"box":[0.22708334028720856,0.8009259104728699,0.34687501192092896,0.970370352268219],"occluded":false},{"timestamp":10.14,"box":[0.24947917461395264,0.9416666626930237,0.27916667610406876,0.9712962917983532],"occluded":false}],"scores":[0.675126850605011,0.675126850605011,0.675126850605011,0.675126850605011,0.675126850605011,0.675126850605011,0.675126850605011,0.675126850605011,0.675126850605011,0.675126850605011]},{"noun_phrase":"wooden beam","object_id":4,"observations":[{"timestamp":7.8,"box":[0.38489586114883423,0.5916666388511658,0.48229169845581055,0.7824073731899261],"occluded":false},{"timestamp":8.06,"box":[0.38593751192092896,0.5351851582527161,0.4880208522081375,0.7166666388511658],"occluded":false},{"timestamp":8.32,"box":[0.40312501788139343,0.5074074268341064,0.5036458596587181,0.7027777880430222],"occluded":false},{"timestamp":8.58,"box":[0.40572917461395264,0.5009258985519409,0.5052083432674408,0.6898147910833359],"occluded":false},{"timestamp":8.84,"box":[0.40416669845581055,0.5083333253860474,0.5052083730697632,0.7009259164333344],"occluded":false},{"timestamp":9.1,"box":[0.39375001192092896,0.5277777910232544,0.49687501788139343,0.7231481522321701],"occluded":false},{"timestamp":9.36,"box":[0.39375001192092896,0.5546296238899231,0.4963541850447655,0.7564814686775208],"occluded":false},{"timestamp":9.620000000000001,"box":[0.40572917461395264,0.6185185313224792,0.5093750134110451,0.85092593729496],"occluded":false},{"timestamp":9.88,"box":[0.21302084624767303,0.7555555701255798,0.5296875387430191,0.9694444537162781],"occluded":false},{"timestamp":10.14,"box":[0.26875001192092896,0.8861110806465149,0.5239583551883698,0.9712962657213211],"occluded":false}],"scores":[0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349]},{"noun_phrase":"wooden beam","object_id":5,"observations":[{"timestamp":8.06,"box":[0.12968750298023224,0.5333333015441895,0.19687500596046448,0.8074073791503906],"occluded":false},{"timestamp":8.32,"box":[0.15000000596046448,0.4981481432914734,0.21458334475755692,0.8101851642131805],"occluded":false},{"timestamp":8.58,"box":[0.15208333730697632,0.4842592477798462,0.21822917461395264,0.8305555284023285],"occluded":false},{"timestamp":8.84,"box":[0.1484375149011612,0.49351850152015686,0.21666668355464935,0.873148113489151],"occluded":false},{"timestamp":9.1,"box":[0.13645833730697632,0.5120370388031006,0.2031250074505806,0.9101851880550385],"occluded":false},{"timestamp":9.36,"box":[0.13437500596046448,0.539814829826355,0.20260417461395264,0.932407408952713],"occluded":false},{"timestamp":9.620000000000001,"box":[0.004687500186264515,0.6157407164573669,0.21406250912696123,0.970370352268219],"occluded":false},{"timestamp":9.88,"box":[0.008333333767950535,0.7268518209457397,0.22552084270864725,0.9703703373670578],"occluded":false},{"timestamp":10.14,"box":[0.01875000074505806,0.8583332896232605,0.26875000074505806,0.9990740269422531],"occluded":false}],"scores":[0.6640419960021973,0.6640419960021973,0.6640419960021973,0.6640419960021973,0.6640419960021973,0.6640419960021973,0.6640419960021973,0.6640419960021973,0.6640419960021973]}],"tracklets":[{"noun_phrase":"stone wall","object_id":0,"presentation_id":1,"observations":[{"timestamp":0.0,"box":[0.6265625357627869,0.032407406717538834,0.9994792342185974,0.9694443978369236],"occluded":false},{"timestamp":0.26,"box":[0.6765625476837158,0.031481482088565826,0.9994792342185974,0.9703703299164772],"occluded":false},{"timestamp":0.52,"box":[0.6458333730697632,0.03055555559694767,0.9994792342185974,0.9712962601333857],"occluded":false},{"timestamp":0.78,"box":[0.6057292222976685,0.03055555559694767,0.9994792342185974,0.9712962601333857],"occluded":false},{"timestamp":1.04,"box":[0.6156250238418579,0.031481482088565826,0.999479204416275,0.9712962880730629],"occluded":false},{"timestamp":1.3,"box":[0.6270833611488342,0.03055555559694767,0.999479204416275,0.9712962601333857],"occluded":false},{"timestamp":1.56,"box":[0.635937511920929,0.03055555559694767,0.999479204416275,0.9712962601333857],"occluded":false},{"timestamp":1.82,"box":[0.004687500186264515,0.025925925001502037,0.9994792463257909,0.9712963011115789],"occluded":false},{"timestamp":2.08,"box":[0.47708335518836975,0.03055555559694767,0.999479204416275,0.9712962601333857],"occluded":false},{"timestamp":2.34,"box":[0.5541666746139526,0.031481482088565826,0.999479204416275,0.9712962880730629],"occluded":false},{"timestamp":2.6,"box":[0.6270833611488342,0.031481482088565826,0.999479204416275,0.9712962880730629],"occluded":false},{"timestamp":2.86,"box":[0.5328125357627869,0.031481482088565826,0.9994792342185974,0.9722221866250038],"occluded":false},{"timestamp":3.12,"box":[0.40781253576278687,0.031481482088565826,0.9994792342185974,0.9712962880730629],"occluded":false},{"timestamp":3.38,"box":[0.6776041984558105,0.031481482088565826,0.999479204416275,0.9712962880730629],"occluded":false},{"timestamp":3.64,"box":[0.26875001192092896,0.031481482088565826,0.9994792342185974,0.9712962880730629],"occluded":false},{"timestamp":3.9,"box":[0.3661458492279053,0.031481482088565826,0.9994792342185974,0.9703703299164772],"occluded":false},{"timestamp":4.16,"box":[0.6338542103767395,0.03055555559694767,0.9994792342185974,0.9712962601333857],"occluded":false},{"timestamp":4.42,"box":[0.5973958373069763,0.03055555559694767,0.999479204416275,0.9712962601333857],"occluded":false},{"timestamp":4.68,"box":[0.6432291865348816,0.031481482088565826,0.999479204416275,0.9703703299164772],"occluded":false},{"timestamp":4.94,"box":[0.643750011920929,0.03055555559694767,0.999479204416275,0.9712962601333857],"occluded":false},{"timestamp":5.2,"box":[0.6541666984558105,0.03055555559694767,0.999479204416275,0.9712962601333857],"occluded":false},{"timestamp":5.46,"box":[0.5604166984558105,0.03055555559694767,0.9994792342185974,0.9712962601333857],"occluded":false},{"timestamp":5.72,"box":[0.5385416746139526,0.03055555559694767,0.999479204416275,0.9712962601333857],"occluded":false},{"timestamp":5.98,"box":[0.6588541865348816,0.03055555559694767,0.999479204416275,0.9712962601333857],"occluded":false},{"timestamp":6.24,"box":[0.6703125238418579,0.03055555559694767,0.999479204416275,0.9712962601333857],"occluded":false},{"timestamp":6.5,"box":[0.6979166865348816,0.03055555559694767,0.999479204416275,0.9712962601333857],"occluded":false},{"timestamp":6.76,"box":[0.8104166984558105,0.03055555559694767,0.999479204416275,0.9712962601333857],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.8984375596046448,0.03055555559694767,0.9994792342185974,0.9712962601333857],"occluded":false}],"mean_score":0.7928802371025085},{"noun_phrase":"stone wall","object_id":2,"presentation_id":2,"observations":[{"timestamp":2.08,"box":[0.3031250238418579,0.10925925523042679,0.5052083730697632,0.9703703597187996],"occluded":false},{"timestamp":2.34,"box":[0.30156251788139343,0.03703703731298447,0.528645858168602,0.9712962731719017],"occluded":false},{"timestamp":2.6,"box":[0.18541668355464935,0.031481482088565826,0.5604167133569717,0.9712962880730629],"occluded":false},{"timestamp":2.86,"box":[0.18541668355464935,0.031481482088565826,0.5614583641290665,0.9712962880730629],"occluded":false},{"timestamp":3.12,"box":[0.11614584177732468,0.03055555559694767,0.8244792148470879,0.9712962601333857],"occluded":false},{"timestamp":3.38,"box":[0.08958333730697632,0.03055555559694767,0.44479167461395264,0.9712962601333857],"occluded":false},{"timestamp":3.64,"box":[0.00937500037252903,0.031481482088565826,0.4197916928678751,0.9712962880730629],"occluded":false},{"timestamp":3.9,"box":[0.0,0.031481482088565826,0.3994791805744171,0.9712962880730629],"occluded":false},{"timestamp":4.16,"box":[0.0,0.03055555559694767,0.3307291865348816,0.969444403424859],"occluded":false},{"timestamp":4.42,"box":[0.0,0.03055555559694767,0.27760419249534607,0.9703703615814447],"occluded":false},{"timestamp":4.68,"box":[0.0,0.031481482088565826,0.22500000894069672,0.9712962880730629],"occluded":false},{"timestamp":4.94,"box":[0.0,0.031481482088565826,0.22604167461395264,0.9990740790963173],"occluded":false},{"timestamp":5.2,"box":[0.0,0.03055555559694767,0.23593750596046448,0.9712962601333857],"occluded":false},{"timestamp":5.46,"box":[0.0,0.03055555559694767,0.22552084922790527,0.9712962601333857],"occluded":false},{"timestamp":5.72,"box":[0.0,0.03055555559694767,0.22447918355464935,0.9712962601333857],"occluded":false},{"timestamp":5.98,"box":[0.0,0.03055555559694767,0.23125001788139343,0.9712962601333857],"occluded":false},{"timestamp":6.24,"box":[0.0,0.03055555559694767,0.2291666716337204,0.9712962601333857],"occluded":false},{"timestamp":6.5,"box":[0.0,0.03055555559694767,0.23281250894069672,0.9712962601333857],"occluded":false},{"timestamp":6.76,"box":[0.0,0.029629629105329514,0.2786458432674408,0.9712962917983532],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.0,0.03055555559694767,0.28072917461395264,0.9712962601333857],"occluded":false},{"timestamp":7.28,"box":[0.0,0.03055555559694767,0.2593750059604645,0.9712962601333857],"occluded":false},{"timestamp":7.54,"box":[0.0,0.2620370388031006,0.9078125357627869,0.9712963104248047],"occluded":false},{"timestamp":7.8,"box":[0.0,0.4037036895751953,0.8104166984558105,0.9722222089767456],"occluded":false},{"timestamp":8.06,"box":[0.0,0.5157407522201538,0.9494792222976685,0.9712963104248047],"occluded":false},{"timestamp":8.32,"box":[0.0,0.48055553436279297,0.2239583432674408,0.9712962806224823],"occluded":false},{"timestamp":8.58,"box":[0.0,0.5388888716697693,0.21822917461395264,0.9712962806224823],"occluded":false},{"timestamp":8.84,"box":[0.0,0.6037036776542664,0.20937500894069672,0.9712962508201599],"occluded":false},{"timestamp":9.1,"box":[0.0,0.6638888716697693,0.1979166716337204,0.9712962806224823],"occluded":false},{"timestamp":9.36,"box":[0.0,0.7064814567565918,0.19062501192092896,0.9712962806224823],"occluded":false},{"timestamp":9.620000000000001,"box":[0.0,0.08148147910833359,0.18593750894069672,0.9712962731719017],"occluded":false},{"timestamp":9.88,"box":[0.0,0.7074074149131775,0.05312500149011612,0.972222238779068],"occluded":false},{"timestamp":10.14,"box":[0.0,0.7694444060325623,0.05781250447034836,0.9722221791744232],"occluded":false},{"timestamp":10.66,"box":[0.0,0.47962963581085205,0.03020833432674408,0.9703703820705414],"occluded":false},{"timestamp":10.92,"box":[0.0,0.032407406717538834,0.07083334028720856,0.9722222127020359],"occluded":false},{"timestamp":11.18,"box":[0.0,0.03055555559694767,0.07916667312383652,0.9518518541008234],"occluded":false},{"timestamp":11.44,"box":[0.0,0.03055555559694767,0.0859375074505806,0.9722222182899714],"occluded":false},{"timestamp":11.700000000000001,"box":[0.0,0.03055555559694767,0.12447917461395264,0.9722222182899714],"occluded":false},{"timestamp":11.96,"box":[0.0,0.03055555559694767,0.15156251192092896,0.9722222182899714],"occluded":false},{"timestamp":12.22,"box":[0.0,0.03055555559694767,0.17395834624767303,0.9722222182899714],"occluded":false},{"timestamp":12.48,"box":[0.0,0.031481482088565826,0.17343750596046448,0.9722221866250038],"occluded":false},{"timestamp":12.74,"box":[0.0,0.03055555559694767,0.16354167461395264,0.9712962601333857],"occluded":false},{"timestamp":13.0,"box":[0.0,0.03055555559694767,0.15572917461395264,0.9712962601333857],"occluded":false},{"timestamp":13.26,"box":[0.0,0.03055555559694767,0.15989583730697632,0.9712962601333857],"occluded":false},{"timestamp":13.52,"box":[0.0,0.03055555559694767,0.16875000298023224,0.9712962601333857],"occluded":false},{"timestamp":13.780000000000001,"box":[0.0,0.03055555559694767,0.17135417461395264,0.9712962601333857],"occluded":false},{"timestamp":14.040000000000001,"box":[0.0,0.03055555559694767,0.17135417461395264,0.9712962601333857],"occluded":false},{"timestamp":14.3,"box":[0.0,0.03055555559694767,0.1979166716337204,0.9712962601333857],"occluded":false},{"timestamp":14.56,"box":[0.0,0.03055555559694767,0.19166667759418488,0.9712962601333857],"occluded":false},{"timestamp":14.82,"box":[0.0,0.03055555559694767,0.15468750894069672,0.7685185167938471],"occluded":false},{"timestamp":15.08,"box":[0.0,0.029629629105329514,0.10364583879709244,0.9703703336417675],"occluded":false}],"mean_score":0.6640419960021973},{"noun_phrase":"wooden beam","object_id":2,"presentation_id":3,"observations":[{"timestamp":7.28,"box":[0.32604169845581055,0.6370370388031006,0.8020834028720856,0.7212962955236435],"occluded":false},{"timestamp":7.54,"box":[0.31041666865348816,0.5537036657333374,0.788541704416275,0.6425925493240356],"occluded":false},{"timestamp":7.8,"box":[0.2994791865348816,0.5314814448356628,0.8109375238418579,0.623148113489151],"occluded":false},{"timestamp":8.06,"box":[0.2291666716337204,0.46666666865348816,0.8250000327825546,0.565740741789341],"occluded":false},{"timestamp":8.32,"box":[0.25468751788139343,0.4546296298503876,0.858854204416275,0.6574074029922485],"occluded":false},{"timestamp":8.58,"box":[0.2828125059604645,0.45092591643333435,0.8718750178813934,0.6574073880910873],"occluded":false},{"timestamp":8.84,"box":[0.29635417461395264,0.4592592418193817,0.8723958730697632,0.6712962687015533],"occluded":false},{"timestamp":9.1,"box":[0.26875001192092896,0.4740740656852722,0.8604167103767395,0.6574074029922485],"occluded":false},{"timestamp":9.36,"box":[0.31197917461395264,0.4962962865829468,0.8572916984558105,0.5935185104608536],"occluded":false},{"timestamp":9.620000000000001,"box":[0.3369791805744171,0.5722222328186035,0.8750000298023224,0.6722222343087196],"occluded":false},{"timestamp":9.88,"box":[0.34322917461395264,0.7037037014961243,0.9041666984558105,0.8074074015021324],"occluded":false},{"timestamp":10.14,"box":[0.3671875298023224,0.8481481075286865,0.9218750298023224,0.9490740299224854],"occluded":false}],"mean_score":0.6559139490127563},{"noun_phrase":"stone wall","object_id":1,"presentation_id":4,"observations":[{"timestamp":0.0,"box":[0.0,0.031481482088565826,0.21145834028720856,0.21388889104127884],"occluded":false},{"timestamp":0.26,"box":[0.06145833805203438,0.03055555559694767,0.2604166753590107,0.23425925709307194],"occluded":false},{"timestamp":0.52,"box":[0.0,0.03055555559694767,0.19739584624767303,0.1999999973922968],"occluded":false},{"timestamp":0.78,"box":[0.0,0.03055555559694767,0.12968750298023224,0.15277777798473835],"occluded":false},{"timestamp":1.04,"box":[0.0,0.03055555559694767,0.11666667461395264,0.1305555570870638],"occluded":false},{"timestamp":1.3,"box":[0.0,0.03055555559694767,0.30000001192092896,0.17962963320314884],"occluded":false},{"timestamp":1.56,"box":[0.0,0.031481482088565826,0.0078125,0.0842592604458332],"occluded":false}],"mean_score":0.6878048777580261},{"noun_phrase":"wooden beam","object_id":3,"presentation_id":5,"observations":[{"timestamp":7.8,"box":[0.18281251192092896,0.6435185074806213,0.30625002086162567,0.8009259104728699],"occluded":false},{"timestamp":8.06,"box":[0.1875000149011612,0.585185170173645,0.30781251937150955,0.7388888746500015],"occluded":false},{"timestamp":8.32,"box":[0.2031250149011612,0.5435184836387634,0.35572919249534607,0.7074073702096939],"occluded":false},{"timestamp":8.58,"box":[0.20520834624767303,0.5324074029922485,0.32760418951511383,0.6990740597248077],"occluded":false},{"timestamp":8.84,"box":[0.20260417461395264,0.550000011920929,0.3244791775941849,0.7074074149131775],"occluded":false},{"timestamp":9.1,"box":[0.19114583730697632,0.565740704536438,0.3166666775941849,0.7296295911073685],"occluded":false},{"timestamp":9.36,"box":[0.18906250596046448,0.5972222089767456,0.31562501192092896,0.7564814686775208],"occluded":false},{"timestamp":9.620000000000001,"box":[0.2083333432674408,0.6777777671813965,0.3291666805744171,0.8333333134651184],"occluded":false},{"timestamp":9.88,"box":[0.22708334028720856,0.8009259104728699,0.34687501192092896,0.970370352268219],"occluded":false},{"timestamp":10.14,"box":[0.24947917461395264,0.9416666626930237,0.27916667610406876,0.9712962917983532],"occluded":false}],"mean_score":0.675126850605011},{"noun_phrase":"wooden beam","object_id":4,"presentation_id":6,"observations":[{"timestamp":7.8,"box":[0.38489586114883423,0.5916666388511658,0.48229169845581055,0.7824073731899261],"occluded":false},{"timestamp":8.06,"box":[0.38593751192092896,0.5351851582527161,0.4880208522081375,0.7166666388511658],"occluded":false},{"timestamp":8.32,"box":[0.40312501788139343,0.5074074268341064,0.5036458596587181,0.7027777880430222],"occluded":false},{"timestamp":8.58,"box":[0.40572917461395264,0.5009258985519409,0.5052083432674408,0.6898147910833359],"occluded":false},{"timestamp":8.84,"box":[0.40416669845581055,0.5083333253860474,0.5052083730697632,0.7009259164333344],"occluded":false},{"timestamp":9.1,"box":[0.39375001192092896,0.5277777910232544,0.49687501788139343,0.7231481522321701],"occluded":false},{"timestamp":9.36,"box":[0.39375001192092896,0.5546296238899231,0.4963541850447655,0.7564814686775208],"occluded":false},{"timestamp":9.620000000000001,"box":[0.40572917461395264,0.6185185313224792,0.5093750134110451,0.85092593729496],"occluded":false},{"timestamp":9.88,"box":[0.21302084624767303,0.7555555701255798,0.5296875387430191,0.9694444537162781],"occluded":false},{"timestamp":10.14,"box":[0.26875001192092896,0.8861110806465149,0.5239583551883698,0.9712962657213211],"occluded":false}],"mean_score":0.6521739363670349},{"noun_phrase":"wooden beam","object_id":5,"presentation_id":7,"observations":[{"timestamp":8.06,"box":[0.12968750298023224,0.5333333015441895,0.19687500596046448,0.8074073791503906],"occluded":false},{"timestamp":8.32,"box":[0.15000000596046448,0.4981481432914734,0.21458334475755692,0.8101851642131805],"occluded":false},{"timestamp":8.58,"box":[0.15208333730697632,0.4842592477798462,0.21822917461395264,0.8305555284023285],"occluded":false},{"timestamp":8.84,"box":[0.1484375149011612,0.49351850152015686,0.21666668355464935,0.873148113489151],"occluded":false},{"timestamp":9.1,"box":[0.13645833730697632,0.5120370388031006,0.2031250074505806,0.9101851880550385],"occluded":false},{"timestamp":9.36,"box":[0.13437500596046448,0.539814829826355,0.20260417461395264,0.932407408952713],"occluded":false},{"timestamp":9.620000000000001,"box":[0.004687500186264515,0.6157407164573669,0.21406250912696123,0.970370352268219],"occluded":false},{"timestamp":9.88,"box":[0.008333333767950535,0.7268518209457397,0.22552084270864725,0.9703703373670578],"occluded":false},{"timestamp":10.14,"box":[0.01875000074505806,0.8583332896232605,0.26875000074505806,0.9990740269422531],"occluded":false}],"mean_score":0.6640419960021973},{"noun_phrase":"wooden beam","object_id":1,"presentation_id":8,"observations":[{"timestamp":5.72,"box":[0.5135416984558105,0.03055555559694767,0.5755208693444729,0.6305555198341608],"occluded":false},{"timestamp":5.98,"box":[0.5197917222976685,0.03055555559694767,0.5786458924412727,0.6407407019287348],"occluded":false},{"timestamp":6.24,"box":[0.5223958492279053,0.03055555559694767,0.5828125178813934,0.6564814541488886],"occluded":false},{"timestamp":6.5,"box":[0.5354167222976685,0.03055555559694767,0.5953125581145287,0.6518518421798944],"occluded":false},{"timestamp":6.76,"box":[0.5953125357627869,0.03055555559694767,0.6453125402331352,0.6435185167938471],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.6244791746139526,0.03055555559694767,0.6755208447575569,0.5601852033287287],"occluded":false},{"timestamp":7.28,"box":[0.6395833492279053,0.031481482088565826,0.6958333514630795,0.5120370164513588],"occluded":false},{"timestamp":7.54,"box":[0.6291667222976685,0.031481482088565826,0.7828125655651093,0.5740740671753883],"occluded":false},{"timestamp":7.8,"box":[0.6338542103767395,0.031481482088565826,0.7968750596046448,0.5879629626870155],"occluded":false},{"timestamp":8.06,"box":[0.643750011920929,0.03055555559694767,0.6958333477377892,0.5324074123054743],"occluded":false},{"timestamp":8.32,"box":[0.31041666865348816,0.03055555559694767,0.721875011920929,0.6574074123054743],"occluded":false},{"timestamp":8.58,"box":[0.2828125059604645,0.03055555559694767,0.7312500178813934,0.6574074123054743],"occluded":false},{"timestamp":8.84,"box":[0.29635417461395264,0.031481482088565826,0.7328125238418579,0.6574073806405067],"occluded":false},{"timestamp":9.1,"box":[0.26875001192092896,0.03055555559694767,0.7187500298023224,0.6990740392357111],"occluded":false},{"timestamp":9.36,"box":[0.31041666865348816,0.031481482088565826,0.7177083492279053,0.7268518581986427],"occluded":false},{"timestamp":9.620000000000001,"box":[0.29635417461395264,0.03055555559694767,0.7322916984558105,0.7962963078171015],"occluded":false},{"timestamp":9.88,"box":[0.32447919249534607,0.031481482088565826,0.7588542103767395,0.9351851716637611],"occluded":false},{"timestamp":10.14,"box":[0.7109375596046448,0.031481482088565826,0.7567708939313889,0.6898147836327553],"occluded":false},{"timestamp":10.4,"box":[0.01875000074505806,0.029629629105329514,0.9635417349636555,0.9990740828216076],"occluded":false},{"timestamp":10.66,"box":[0.06041666865348816,0.03055555559694767,0.9911458790302277,0.99907405115664],"occluded":false},{"timestamp":10.92,"box":[0.03281250223517418,0.03055555559694767,0.9911458753049374,0.99907405115664],"occluded":false},{"timestamp":11.18,"box":[0.03281250223517418,0.031481482088565826,0.9494791887700558,0.9990740790963173],"occluded":false},{"timestamp":11.44,"box":[0.07447917014360428,0.032407406717538834,0.9911458566784859,0.9990740455687046],"occluded":false},{"timestamp":11.700000000000001,"box":[0.33802086114883423,0.031481482088565826,0.99114590883255,0.9990740790963173],"occluded":false},{"timestamp":11.96,"box":[0.7385417222976685,0.043518517166376114,0.8270833939313889,0.9712962843477726],"occluded":false},{"timestamp":12.22,"box":[0.3802083432674408,0.11759258806705475,0.8296875357627869,0.9842592626810074],"occluded":false},{"timestamp":12.48,"box":[0.5604166984558105,0.21851851046085358,0.8223958909511566,0.9990740567445755],"occluded":false},{"timestamp":12.74,"box":[0.2828125059604645,0.3092592656612396,0.807291716337204,0.9990740716457367],"occluded":false},{"timestamp":13.0,"box":[0.15781250596046448,0.37870368361473083,0.7947917282581329,0.9990740716457367],"occluded":false},{"timestamp":13.26,"box":[0.24114584922790527,0.4342592656612396,0.7843750715255737,0.9990740716457367],"occluded":false},{"timestamp":13.52,"box":[0.4489583671092987,0.4824073910713196,0.776041716337204,0.9990740418434143],"occluded":false},{"timestamp":13.780000000000001,"box":[0.3661458492279053,0.5074074268341064,0.7614583671092987,0.9990740716457367],"occluded":false},{"timestamp":14.040000000000001,"box":[0.26875001192092896,0.5166666507720947,0.7536458671092987,0.9990740418434143],"occluded":false},{"timestamp":14.3,"box":[0.6822916865348816,0.5324074029922485,0.747395858168602,0.970370352268219],"occluded":false},{"timestamp":14.56,"box":[0.6562500596046448,0.5462962985038757,0.7104167304933071,0.970370352268219],"occluded":false},{"timestamp":14.82,"box":[0.6020833849906921,0.5324074029922485,0.6421875543892384,0.9712962806224823],"occluded":false},{"timestamp":15.08,"box":[0.15781250596046448,0.046296294778585434,0.5687500238418579,0.9703703485429287],"occluded":false},{"timestamp":15.34,"box":[0.32447919249534607,0.4435185194015503,0.40781252831220627,0.9712963104248047],"occluded":false}],"mean_score":0.6995305418968201},{"noun_phrase":"wooden beam","object_id":0,"presentation_id":9,"observations":[{"timestamp":5.72,"box":[0.2031250149011612,0.575925886631012,0.5057291835546494,0.6083332933485508],"occluded":false},{"timestamp":5.98,"box":[0.21197918057441711,0.5805555582046509,0.5125000178813934,0.6111111138015985],"occluded":false},{"timestamp":6.24,"box":[0.20625001192092896,0.5972222089767456,0.5140625238418579,0.6268518380820751],"occluded":false},{"timestamp":6.5,"box":[0.2057291716337204,0.5907407402992249,0.5244791954755783,0.6222222223877907],"occluded":false},{"timestamp":6.76,"box":[0.2421875149011612,0.5861111283302307,0.583333358168602,0.6148148309439421],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.21250000596046448,0.5018518567085266,0.6119791865348816,0.5490740798413754],"occluded":false},{"timestamp":7.28,"box":[0.28125,0.44999998807907104,0.6255208551883698,0.4870370253920555],"occluded":false},{"timestamp":7.54,"box":[0.2515625059604645,0.3611111044883728,0.7828125059604645,0.5740740597248077],"occluded":false},{"timestamp":7.8,"box":[0.23229168355464935,0.3305555582046509,0.7828125208616257,0.5879629552364349],"occluded":false},{"timestamp":8.06,"box":[0.22812500596046448,0.2620370388031006,0.6307291984558105,0.31203703954815865],"occluded":false},{"timestamp":8.32,"box":[0.25052085518836975,0.2342592477798462,0.6546875536441803,0.2805555425584316],"occluded":false},{"timestamp":8.58,"box":[0.24114584922790527,0.22499999403953552,0.6604166924953461,0.26759258657693863],"occluded":false},{"timestamp":8.84,"box":[0.24427084624767303,0.23148147761821747,0.6604167073965073,0.27222221717238426],"occluded":false},{"timestamp":9.1,"box":[0.24062500894069672,0.2481481432914734,0.6468750387430191,0.2935185134410858],"occluded":false},{"timestamp":9.36,"box":[0.23229168355464935,0.2759259343147278,0.6442708820104599,0.32407408207654953],"occluded":false},{"timestamp":9.620000000000001,"box":[0.24635417759418488,0.3444444239139557,0.6562500447034836,0.3990740515291691],"occluded":false},{"timestamp":9.88,"box":[0.29479166865348816,0.46296295523643494,0.6994791924953461,0.7546296119689941],"occluded":false},{"timestamp":10.14,"box":[0.30937501788139343,0.5740740895271301,0.6859375238418579,0.6259259395301342],"occluded":false},{"timestamp":10.4,"box":[0.0052083334885537624,0.6962962746620178,0.963541706558317,0.9990740418434143],"occluded":false},{"timestamp":10.66,"box":[0.06041666865348816,0.7685185074806213,0.9494791924953461,0.9990740567445755],"occluded":false},{"timestamp":10.92,"box":[0.06041666865348816,0.032407406717538834,0.8104167282581329,0.9990740455687046],"occluded":false},{"timestamp":11.18,"box":[0.03281250223517418,0.8305555582046509,0.8802083991467953,0.9990740716457367],"occluded":false},{"timestamp":11.44,"box":[0.29635417461395264,0.8638888597488403,0.9078125357627869,0.9990740418434143],"occluded":false},{"timestamp":11.700000000000001,"box":[0.26875001192092896,0.8925925493240356,0.7677083611488342,0.9990740269422531],"occluded":false},{"timestamp":11.96,"box":[0.3369791805744171,0.9351851940155029,0.7723958790302277,0.9722222313284874],"occluded":false}],"mean_score":0.6683937907218933}],"dropped":{},"fates":[{"noun_phrase":"stone wall","object_id":0,"frames":28,"mean_score":0.7928802371025085,"mean_area":0.39063214249721606,"outcome":"kept","presentation_id":1,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"stone wall","object_id":1,"frames":7,"mean_score":0.6878048777580261,"mean_area":0.026457094372017313,"outcome":"kept","presentation_id":4,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"stone wall","object_id":2,"frames":50,"mean_score":0.6640419960021973,"mean_area":0.2075277259586998,"outcome":"kept","presentation_id":2,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"wooden beam","object_id":0,"frames":25,"mean_score":0.6683937907218933,"mean_area":0.08484512166832849,"outcome":"kept","presentation_id":9,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"wooden beam","object_id":1,"frames":38,"mean_score":0.6995305418968201,"mean_area":0.27806462786694136,"outcome":"kept","presentation_id":8,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"wooden beam","object_id":2,"frames":12,"mean_score":0.6559139490127563,"mean_area":0.07368791135002128,"outcome":"kept","presentation_id":3,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"wooden beam","object_id":3,"frames":10,"mean_score":0.675126850605011,"mean_area":0.018321759794748156,"outcome":"kept","presentation_id":5,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"wooden beam","object_id":4,"frames":10,"mean_score":0.6521739363670349,"mean_area":0.02494106974597028,"outcome":"kept","presentation_id":6,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"wooden beam","object_id":5,"frames":9,"mean_score":0.6640419960021973,"mean_area":0.03366989332143297,"outcome":"kept","presentation_id":7,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null}],"gates":{"min_track_frames":3,"min_mean_score":0.5,"merge_iou":0.8,"max_tracklets":12}}},"render":{"inputs":{"tracklets":9,"as_video":true,"fps":1.0,"max_frames":30},"outputs":{"media":"/home/zzk/data/video_bbox/vlm_gemini3.5_lite/renders/--cif7Nv6BI_chunk_014_scene_002_choices.mp4","media_bytes":4277923,"frames":[{"index":1,"timestamp":0.0,"file":"--cif7Nv6BI_chunk_014_scene_002_choices.mp4"},{"index":2,"timestamp":1.04,"file":"--cif7Nv6BI_chunk_014_scene_002_choices.mp4"},{"index":3,"timestamp":2.08,"file":"--cif7Nv6BI_chunk_014_scene_002_choices.mp4"},{"index":4,"timestamp":3.12,"file":"--cif7Nv6BI_chunk_014_scene_002_choices.mp4"},{"index":5,"timestamp":4.16,"file":"--cif7Nv6BI_chunk_014_scene_002_choices.mp4"},{"index":6,"timestamp":5.2,"file":"--cif7Nv6BI_chunk_014_scene_002_choices.mp4"},{"index":7,"timestamp":6.24,"file":"--cif7Nv6BI_chunk_014_scene_002_choices.mp4"},{"index":8,"timestamp":7.28,"file":"--cif7Nv6BI_chunk_014_scene_002_choices.mp4"},{"index":9,"timestamp":8.06,"file":"--cif7Nv6BI_chunk_014_scene_002_choices.mp4"},{"index":10,"timestamp":9.1,"file":"--cif7Nv6BI_chunk_014_scene_002_choices.mp4"},{"index":11,"timestamp":10.14,"file":"--cif7Nv6BI_chunk_014_scene_002_choices.mp4"},{"index":12,"timestamp":11.18,"file":"--cif7Nv6BI_chunk_014_scene_002_choices.mp4"},{"index":13,"timestamp":12.22,"file":"--cif7Nv6BI_chunk_014_scene_002_choices.mp4"},{"index":14,"timestamp":13.26,"file":"--cif7Nv6BI_chunk_014_scene_002_choices.mp4"},{"index":15,"timestamp":14.3,"file":"--cif7Nv6BI_chunk_014_scene_002_choices.mp4"},{"index":16,"timestamp":15.34,"file":"--cif7Nv6BI_chunk_014_scene_002_choices.mp4"}]}},"select":{"inputs":{"model":"gemini-3.5-flash-lite","temperature":0.0,"as_video":true,"media_bytes":4277923,"media":["--cif7Nv6BI_chunk_014_scene_002_choices.mp4"],"choice_frames":16,"system_prompt":"You are labelling video data for referring multi-object tracking.\n\nYou will see a numbered sequence of frames from one clip, in time order. Each frame has its own number and timestamp printed in the top-left corner. Candidate objects have been detected for you and drawn on the frames as coloured boxes, each labelled with a number. The same object keeps the same number and the same colour on every frame. Below each frame is a strip of enlarged crops of those same numbered boxes, so you can identify a small or distant object.\n\nFor each description you are given, answer two questions:\n\n1. MATCHING_IDS — which of the numbered objects the description refers to. List every number that matches, or an EMPTY LIST if none of them do. An empty list is a correct and expected answer: some descriptions are deliberately about something that is not in this clip. Do not pick a number just to have picked one, and do not pick a number for an object that merely resembles the description.\n\n2. START_FRAME and END_FRAME — the frame numbers between which the described state actually holds. If the description is \"the car that's turning left\", the interval is the frames in which that car is turning, NOT every frame in which the car is visible. Use the frame numbers printed on the frames. START_FRAME must not be greater than END_FRAME. When MATCHING_IDS is empty, give the full range (the first and last frame numbers); it is ignored.\n\n3. CONFIDENCE — how sure you are of the MATCHING_IDS answer, from 0.0 to 1.0. Be honest: a low number costs the example nothing, and an overconfident wrong answer becomes a wrong label.\n\nRules:\n- Only use numbers that are actually drawn on the frames.\n- Never output coordinates, box values, or pixel positions.\n- Judge motion by comparing the object's position across frames, not by guessing from one frame.\n","user_prompt":"The video has 16 frames, numbered 1 to 16. Each frame is one moment of the clip and has its number burned into the top left corner.\n\nAnswer for each of the following descriptions, in the order given:\n1. the stone wall illuminated by the light\n2. the wooden beams of the ruined ceiling\n3. the modern metal railing","questions":["the stone wall illuminated by the light","the wooden beams of the ruined ceiling","the modern metal railing"]},"outputs":{"raw_response":"{\n  \"answers\": [\n    {\n      \"description_index\": 1,\n      \"matching_ids\": [\n        1\n      ],\n      \"start_frame\": 1,\n      \"end_frame\": 16,\n      \"confidence\": 1.0\n    },\n    {\n      \"description_index\": 2,\n      \"matching_ids\": [\n        3,\n        8\n      ],\n      \"start_frame\": 7,\n      \"end_frame\": 16,\n      \"confidence\": 0.9\n    },\n    {\n      \"description_index\": 3,\n      \"matching_ids\": [],\n      \"start_frame\": 1,\n      \"end_frame\": 16,\n      \"confidence\": 1.0\n    }\n  ]\n}","raw_answers":[{"description_index":1,"matching_ids":[1],"start_frame":1,"end_frame":16,"confidence":1.0},{"description_index":2,"matching_ids":[3,8],"start_frame":7,"end_frame":16,"confidence":0.9},{"description_index":3,"matching_ids":[],"start_frame":1,"end_frame":16,"confidence":1.0}],"answers":[{"expression":"the stone wall illuminated by the light","matching_presentation_ids":[1],"start_choice_frame":1,"end_choice_frame":16,"confidence":1.0,"self_reported_confidence":1.0},{"expression":"the wooden beams of the ruined ceiling","matching_presentation_ids":[3,8],"start_choice_frame":7,"end_choice_frame":16,"confidence":0.9,"self_reported_confidence":0.9},{"expression":"the modern metal railing","matching_presentation_ids":[],"start_choice_frame":1,"end_choice_frame":16,"confidence":1.0,"self_reported_confidence":1.0}]}},"verify":{"inputs":{"expressions":3,"tracklets":9,"selections":3,"gates":"interval, count, unknown ids, confidence, then geometry (motion)"},"outputs":{"kept":[{"expression":"the stone wall illuminated by the light","kind":"appearance","is_negative":false,"tracks":[{"timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005],"boxes":[[0.6265625357627869,0.032407406717538834,0.9994792342185974,0.9694443978369236],[0.6765625476837158,0.031481482088565826,0.9994792342185974,0.9703703299164772],[0.6458333730697632,0.03055555559694767,0.9994792342185974,0.9712962601333857],[0.6057292222976685,0.03055555559694767,0.9994792342185974,0.9712962601333857],[0.6156250238418579,0.031481482088565826,0.999479204416275,0.9712962880730629],[0.6270833611488342,0.03055555559694767,0.999479204416275,0.9712962601333857],[0.635937511920929,0.03055555559694767,0.999479204416275,0.9712962601333857],[0.004687500186264515,0.025925925001502037,0.9994792463257909,0.9712963011115789],[0.47708335518836975,0.03055555559694767,0.999479204416275,0.9712962601333857],[0.5541666746139526,0.031481482088565826,0.999479204416275,0.9712962880730629],[0.6270833611488342,0.031481482088565826,0.999479204416275,0.9712962880730629],[0.5328125357627869,0.031481482088565826,0.9994792342185974,0.9722221866250038],[0.40781253576278687,0.031481482088565826,0.9994792342185974,0.9712962880730629],[0.6776041984558105,0.031481482088565826,0.999479204416275,0.9712962880730629],[0.26875001192092896,0.031481482088565826,0.9994792342185974,0.9712962880730629],[0.3661458492279053,0.031481482088565826,0.9994792342185974,0.9703703299164772],[0.6338542103767395,0.03055555559694767,0.9994792342185974,0.9712962601333857],[0.5973958373069763,0.03055555559694767,0.999479204416275,0.9712962601333857],[0.6432291865348816,0.031481482088565826,0.999479204416275,0.9703703299164772],[0.643750011920929,0.03055555559694767,0.999479204416275,0.9712962601333857],[0.6541666984558105,0.03055555559694767,0.999479204416275,0.9712962601333857],[0.5604166984558105,0.03055555559694767,0.9994792342185974,0.9712962601333857],[0.5385416746139526,0.03055555559694767,0.999479204416275,0.9712962601333857],[0.6588541865348816,0.03055555559694767,0.999479204416275,0.9712962601333857],[0.6703125238418579,0.03055555559694767,0.999479204416275,0.9712962601333857],[0.6979166865348816,0.03055555559694767,0.999479204416275,0.9712962601333857],[0.8104166984558105,0.03055555559694767,0.999479204416275,0.9712962601333857],[0.8984375596046448,0.03055555559694767,0.9994792342185974,0.9712962601333857]]}]}],"rejected":{"the modern metal railing":"no_candidates_retrieved","the wooden beams of the ruined ceiling":"count_mismatch"},"rejection_counts":{"no_candidates_retrieved":1,"count_mismatch":1},"row":{"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_014_scene_002.mp4","scene_key":"--cif7Nv6BI_chunk_014_scene_002","category":"test_vehicles","height":1080,"width":1920,"fps":50.0,"num_frames":770,"duration":15.4,"rmot_expressions":[{"expression":"the stone wall illuminated by the light","kind":"appearance","is_negative":false,"tracks":[{"timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005],"boxes":[[0.6265625357627869,0.032407406717538834,0.9994792342185974,0.9694443978369236],[0.6765625476837158,0.031481482088565826,0.9994792342185974,0.9703703299164772],[0.6458333730697632,0.03055555559694767,0.9994792342185974,0.9712962601333857],[0.6057292222976685,0.03055555559694767,0.9994792342185974,0.9712962601333857],[0.6156250238418579,0.031481482088565826,0.999479204416275,0.9712962880730629],[0.6270833611488342,0.03055555559694767,0.999479204416275,0.9712962601333857],[0.635937511920929,0.03055555559694767,0.999479204416275,0.9712962601333857],[0.004687500186264515,0.025925925001502037,0.9994792463257909,0.9712963011115789],[0.47708335518836975,0.03055555559694767,0.999479204416275,0.9712962601333857],[0.5541666746139526,0.031481482088565826,0.999479204416275,0.9712962880730629],[0.6270833611488342,0.031481482088565826,0.999479204416275,0.9712962880730629],[0.5328125357627869,0.031481482088565826,0.9994792342185974,0.9722221866250038],[0.40781253576278687,0.031481482088565826,0.9994792342185974,0.9712962880730629],[0.6776041984558105,0.031481482088565826,0.999479204416275,0.9712962880730629],[0.26875001192092896,0.031481482088565826,0.9994792342185974,0.9712962880730629],[0.3661458492279053,0.031481482088565826,0.9994792342185974,0.9703703299164772],[0.6338542103767395,0.03055555559694767,0.9994792342185974,0.9712962601333857],[0.5973958373069763,0.03055555559694767,0.999479204416275,0.9712962601333857],[0.6432291865348816,0.031481482088565826,0.999479204416275,0.9703703299164772],[0.643750011920929,0.03055555559694767,0.999479204416275,0.9712962601333857],[0.6541666984558105,0.03055555559694767,0.999479204416275,0.9712962601333857],[0.5604166984558105,0.03055555559694767,0.9994792342185974,0.9712962601333857],[0.5385416746139526,0.03055555559694767,0.999479204416275,0.9712962601333857],[0.6588541865348816,0.03055555559694767,0.999479204416275,0.9712962601333857],[0.6703125238418579,0.03055555559694767,0.999479204416275,0.9712962601333857],[0.6979166865348816,0.03055555559694767,0.999479204416275,0.9712962601333857],[0.8104166984558105,0.03055555559694767,0.999479204416275,0.9712962601333857],[0.8984375596046448,0.03055555559694767,0.9994792342185974,0.9712962601333857]]}]}],"rmot_expression_count":1,"rmot_negative_count":0,"rmot_motion_count":0}}}}},{"index":4,"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_008_scene_014.mp4","status":"complete","steps":{"decode":{"inputs":{"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_008_scene_014.mp4","sample_fps":4.0},"outputs":{"width":1920,"height":1080,"source_fps":50.0,"duration_seconds":11.88,"frames_kept":46,"blank_frames_dropped":0,"timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88,10.14,10.4,10.66,10.92,11.18,11.44,11.700000000000001],"error":""}},"propose":{"inputs":{"model":"gemini-3.5-flash-lite","temperature":0.4,"video_bytes":15512964,"system_prompt":"You are annotating video clips to build training data for referring multi-object tracking: given a video and a natural-language description, a model must output the boxes of every object matching that description, over time.\n\nYour job is to write the descriptions. You do NOT output coordinates, boxes, pixel positions, or any numbers about where things are. A separate detector finds the objects.\n\nFor every clip you produce a list of items. Each item has two texts that must NOT be the same:\n\nEXPRESSION — what a person would say to single out the object or objects. It may describe motion, temporal order, interaction with other objects, and spatial relations. It should read like something a person would actually say.\n  Good: \"the black car that's turning left\"\n  Good: \"the cyclist who overtakes the bus\"\n  Good: \"the two people who walk out of frame on the right\"\n\nNOUN_PHRASE — the same object as a plain, timeless visual category with its visible attributes, and NOTHING else. The detector is prompted with this text and looks at ONE frame at a time, so anything about time, order, or action is useless to it and makes it fail.\n  Allowed: category nouns, colours, materials, clothing, size, type\n  Forbidden: verbs and participles of action (\"turning\", \"walks\", \"is moving\"), temporal words (\"then\", \"starts\", \"after\", \"first\", \"last\"), relative clauses (\"that ...\", \"which ...\", \"who ...\"), ordinals and rank words (\"first\", \"leftmost\", \"nearest\"), and anything over six words.\n  \"the black car that's turning left\" -> \"black car\"\n  \"the cyclist who overtakes the bus\" -> \"cyclist\"\n  \"the two people who walk out of frame on the right\" -> \"person\"\n\nKIND — what makes the expression hard: \"motion\", \"appearance\", \"spatial\", or \"interaction\". Aim for about 35% \"motion\" items: motion is the part a single-frame model cannot already solve, so it is the most valuable, but a list of only motion items is not usable training data either.\n\nEXPECTED_COUNT — how many distinct objects in the clip match the EXPRESSION. Count carefully; this is checked against the detector's answer, and a wrong count discards the item.\n\nIS_NEGATIVE — at most ONE item may be a negative, and only when you also write at least one non-negative item. A negative is an expression that recombines attributes that ARE present in the clip into a whole that is NOT. If the only black car is moving, \"the parked black car\" is a good negative. Its NOUN_PHRASE must still be the category that IS present (\"black car\"), because that is what makes the absence checkable rather than merely unverified. A negative's EXPECTED_COUNT is 0.\n\nRules that discard an item, so follow them exactly:\n- Never write coordinates, box values, pixel positions, or percentages of the frame.\n- Never let a NOUN_PHRASE contain a verb of action, a temporal word, a relative clause, or an ordinal.\n- Never write an EXPRESSION whose object is not visible in the clip, unless it is marked IS_NEGATIVE.\n- Prefer objects that are clearly visible and identifiable over background clutter.\n- Do not describe the camera or the shot; describe objects in the scene.\n","user_prompt":"Watch this clip and write between 1 and 3 items following the rules. Cover different objects rather than restating one object several ways, and include any negative in the same list.\n"},"outputs":{"raw_response":"{\n  \"items\": [\n    {\n      \"expression\": \"the man standing on the stone path\",\n      \"noun_phrase\": \"man\",\n      \"kind\": \"appearance\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the dark dog running down the hill\",\n      \"noun_phrase\": \"dark dog\",\n      \"kind\": \"motion\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    }\n  ]\n}","raw_items":[{"expression":"the man standing on the stone path","noun_phrase":"man","kind":"appearance","expected_count":1,"is_negative":false},{"expression":"the dark dog running down the hill","noun_phrase":"dark dog","kind":"motion","expected_count":1,"is_negative":false}],"accepted":[{"expression":"the man standing on the stone path","noun_phrase":"man","kind":"appearance","expected_count":1,"is_negative":false},{"expression":"the dark dog running down the hill","noun_phrase":"dark dog","kind":"motion","expected_count":1,"is_negative":false}],"gated":[]}},"track":{"inputs":{"tracker":"Sam31MultiplexTracker","model_id":"/mnt/ceph3/ec/jacky/sam3.1","noun_phrases":["dark dog","man"],"expressions_in":2,"frames":46,"sample_fps":4.0,"frame_size":"1920x1080","timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88,10.14,10.4,10.66,10.92,11.18,11.44,11.700000000000001],"session":"one session over 46 JPEG frames; then per phrase: reset_session, add_prompt(text) on frame 0, propagate_in_video","passes":2},"outputs":{"raw_tracks":[{"noun_phrase":"dark dog","object_id":0,"observations":[{"timestamp":2.86,"box":[0.5057291984558105,0.5185185074806213,0.5213541984558105,0.574999988079071],"occluded":false},{"timestamp":3.12,"box":[0.4828125238418579,0.5,0.498958358541131,0.5537037029862404],"occluded":false},{"timestamp":3.38,"box":[0.4869791865348816,0.4851851761341095,0.5036458540707827,0.5435185097157955],"occluded":false},{"timestamp":3.64,"box":[0.49010419845581055,0.48055553436279297,0.5062500331550837,0.5462962761521339],"occluded":false},{"timestamp":3.9,"box":[0.4911458492279053,0.4888888895511627,0.5098958499729633,0.5500000007450581],"occluded":false},{"timestamp":4.16,"box":[0.47864586114883423,0.4879629611968994,0.5020833630114794,0.5537037029862404],"occluded":false},{"timestamp":4.42,"box":[0.4713541865348816,0.49444442987442017,0.5072916895151138,0.5583333149552345],"occluded":false},{"timestamp":4.68,"box":[0.46302086114883423,0.5027777552604675,0.5098958648741245,0.5675925686955452],"occluded":false},{"timestamp":4.94,"box":[0.4557291865348816,0.5342592597007751,0.5088541880249977,0.6009259223937988],"occluded":false},{"timestamp":5.2,"box":[0.46302086114883423,0.5583333373069763,0.5151041969656944,0.6240740790963173],"occluded":false},{"timestamp":5.46,"box":[0.4723958671092987,0.5509259104728699,0.527083370834589,0.6157407239079475],"occluded":false},{"timestamp":5.72,"box":[0.4661458432674408,0.54537034034729,0.5203125141561031,0.6101851537823677],"occluded":false},{"timestamp":5.98,"box":[0.483854204416275,0.5879629254341125,0.5338542088866234,0.6555555164813995],"occluded":false},{"timestamp":6.24,"box":[0.659375011920929,0.7620370388031006,0.7312500178813934,0.8509259223937988],"occluded":false},{"timestamp":9.620000000000001,"box":[0.5776041746139526,0.3759259283542633,0.8171875178813934,0.9749999940395355],"occluded":false},{"timestamp":10.4,"box":[0.6390625238418579,0.5407407283782959,0.6817708611488342,0.5990740619599819],"occluded":false},{"timestamp":10.66,"box":[0.5213541984558105,0.5222222208976746,0.5671875327825546,0.5787037014961243],"occluded":false},{"timestamp":10.92,"box":[0.49531251192092896,0.5175926089286804,0.5390625149011612,0.5731481648981571],"occluded":false},{"timestamp":11.18,"box":[0.4791666865348816,0.550000011920929,0.5223958566784859,0.6064814925193787],"occluded":false},{"timestamp":11.44,"box":[0.47864586114883423,0.550000011920929,0.5208333618938923,0.6055555678904057],"occluded":false},{"timestamp":11.700000000000001,"box":[0.476041704416275,0.550000011920929,0.5171875394880772,0.6055555678904057],"occluded":false}],"scores":[0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035]},{"noun_phrase":"man","object_id":0,"observations":[{"timestamp":0.0,"box":[0.5973958373069763,0.3638888895511627,0.6526041738688946,0.6416666507720947],"occluded":false},{"timestamp":0.26,"box":[0.5838541984558105,0.385185182094574,0.6385417021811008,0.6462962925434113],"occluded":false},{"timestamp":0.52,"box":[0.5614583492279053,0.40740740299224854,0.6062500178813934,0.6611111164093018],"occluded":false},{"timestamp":0.78,"box":[0.5640625357627869,0.4138888716697693,0.6114583723247051,0.6712962687015533],"occluded":false},{"timestamp":1.04,"box":[0.5281250476837158,0.4268518388271332,0.5791667178273201,0.6925925612449646],"occluded":false},{"timestamp":1.3,"box":[0.5083333849906921,0.4120370149612427,0.5744792222976685,0.680555522441864],"occluded":false},{"timestamp":1.56,"box":[0.4932291805744171,0.4203703701496124,0.5473958514630795,0.6972222030162811],"occluded":false},{"timestamp":1.82,"box":[0.4567708671092987,0.4435185194015503,0.511979203671217,0.7287037074565887],"occluded":false},{"timestamp":2.08,"box":[0.3723958432674408,0.4287036955356598,0.42239584773778915,0.7249999940395355],"occluded":false},{"timestamp":6.76,"box":[0.24062500894069672,0.3712962865829468,0.32083334773778915,0.9990740418434143],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.39375001192092896,0.4148148000240326,0.5010416805744171,0.835185170173645],"occluded":false},{"timestamp":7.28,"box":[0.4791666865348816,0.4314814805984497,0.5546875223517418,0.9277777671813965],"occluded":false},{"timestamp":7.54,"box":[0.5078125,0.46481481194496155,0.5958333387970924,0.972222238779068],"occluded":false},{"timestamp":7.8,"box":[0.5281250476837158,0.432407408952713,0.6140625551342964,0.972222238779068],"occluded":false},{"timestamp":8.06,"box":[0.5338541865348816,0.4148148000240326,0.6312500238418579,0.9712962806224823],"occluded":false},{"timestamp":8.32,"box":[0.5864583849906921,0.4157407283782959,0.6973958909511566,0.9722222089767456],"occluded":false},{"timestamp":8.58,"box":[0.6291667222976685,0.39814814925193787,0.7557292282581329,0.972222238779068],"occluded":false},{"timestamp":8.84,"box":[0.6500000357627869,0.39629629254341125,0.8026042133569717,0.9712962806224823],"occluded":false},{"timestamp":9.1,"box":[0.6328125596046448,0.42222222685813904,0.8234375715255737,0.9749999940395355],"occluded":false},{"timestamp":9.36,"box":[0.6067708730697632,0.3731481432914734,0.8317708820104599,0.9740740656852722],"occluded":false},{"timestamp":9.620000000000001,"box":[0.5802083611488342,0.3740740716457367,0.8166667073965073,0.9759258925914764],"occluded":false},{"timestamp":9.88,"box":[0.24635417759418488,0.2935185134410858,0.46822918951511383,0.9731481373310089],"occluded":false}],"scores":[0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333]}],"tracklets":[{"noun_phrase":"man","object_id":0,"presentation_id":1,"observations":[{"timestamp":0.0,"box":[0.5973958373069763,0.3638888895511627,0.6526041738688946,0.6416666507720947],"occluded":false},{"timestamp":0.26,"box":[0.5838541984558105,0.385185182094574,0.6385417021811008,0.6462962925434113],"occluded":false},{"timestamp":0.52,"box":[0.5614583492279053,0.40740740299224854,0.6062500178813934,0.6611111164093018],"occluded":false},{"timestamp":0.78,"box":[0.5640625357627869,0.4138888716697693,0.6114583723247051,0.6712962687015533],"occluded":false},{"timestamp":1.04,"box":[0.5281250476837158,0.4268518388271332,0.5791667178273201,0.6925925612449646],"occluded":false},{"timestamp":1.3,"box":[0.5083333849906921,0.4120370149612427,0.5744792222976685,0.680555522441864],"occluded":false},{"timestamp":1.56,"box":[0.4932291805744171,0.4203703701496124,0.5473958514630795,0.6972222030162811],"occluded":false},{"timestamp":1.82,"box":[0.4567708671092987,0.4435185194015503,0.511979203671217,0.7287037074565887],"occluded":false},{"timestamp":2.08,"box":[0.3723958432674408,0.4287036955356598,0.42239584773778915,0.7249999940395355],"occluded":false},{"timestamp":6.76,"box":[0.24062500894069672,0.3712962865829468,0.32083334773778915,0.9990740418434143],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.39375001192092896,0.4148148000240326,0.5010416805744171,0.835185170173645],"occluded":false},{"timestamp":7.28,"box":[0.4791666865348816,0.4314814805984497,0.5546875223517418,0.9277777671813965],"occluded":false},{"timestamp":7.54,"box":[0.5078125,0.46481481194496155,0.5958333387970924,0.972222238779068],"occluded":false},{"timestamp":7.8,"box":[0.5281250476837158,0.432407408952713,0.6140625551342964,0.972222238779068],"occluded":false},{"timestamp":8.06,"box":[0.5338541865348816,0.4148148000240326,0.6312500238418579,0.9712962806224823],"occluded":false},{"timestamp":8.32,"box":[0.5864583849906921,0.4157407283782959,0.6973958909511566,0.9722222089767456],"occluded":false},{"timestamp":8.58,"box":[0.6291667222976685,0.39814814925193787,0.7557292282581329,0.972222238779068],"occluded":false},{"timestamp":8.84,"box":[0.6500000357627869,0.39629629254341125,0.8026042133569717,0.9712962806224823],"occluded":false},{"timestamp":9.1,"box":[0.6328125596046448,0.42222222685813904,0.8234375715255737,0.9749999940395355],"occluded":false},{"timestamp":9.36,"box":[0.6067708730697632,0.3731481432914734,0.8317708820104599,0.9740740656852722],"occluded":false},{"timestamp":9.620000000000001,"box":[0.5802083611488342,0.3740740716457367,0.8166667073965073,0.9759258925914764],"occluded":false},{"timestamp":9.88,"box":[0.24635417759418488,0.2935185134410858,0.46822918951511383,0.9731481373310089],"occluded":false}],"mean_score":0.9569892287254333}],"dropped":{"merged_duplicate":1},"fates":[{"noun_phrase":"dark dog","object_id":0,"frames":21,"mean_score":0.7852349281311035,"mean_area":0.009223986185546727,"outcome":"merged_duplicate","presentation_id":null,"merged_into_phrase":"man","merged_into_object_id":0,"merge_iou":0.982460531650171},{"noun_phrase":"man","object_id":0,"frames":22,"mean_score":0.9569892287254333,"mean_area":0.052912416322769266,"outcome":"kept","presentation_id":1,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null}],"gates":{"min_track_frames":3,"min_mean_score":0.5,"merge_iou":0.8,"max_tracklets":12}}},"render":{"inputs":{"tracklets":1,"as_video":true,"fps":1.0,"max_frames":30},"outputs":{"media":"/home/zzk/data/video_bbox/vlm_gemini3.5_lite/renders/--cif7Nv6BI_chunk_008_scene_014_choices.mp4","media_bytes":3587038,"frames":[{"index":1,"timestamp":0.0,"file":"--cif7Nv6BI_chunk_008_scene_014_choices.mp4"},{"index":2,"timestamp":1.04,"file":"--cif7Nv6BI_chunk_008_scene_014_choices.mp4"},{"index":3,"timestamp":2.08,"file":"--cif7Nv6BI_chunk_008_scene_014_choices.mp4"},{"index":4,"timestamp":2.86,"file":"--cif7Nv6BI_chunk_008_scene_014_choices.mp4"},{"index":5,"timestamp":3.9,"file":"--cif7Nv6BI_chunk_008_scene_014_choices.mp4"},{"index":6,"timestamp":4.94,"file":"--cif7Nv6BI_chunk_008_scene_014_choices.mp4"},{"index":7,"timestamp":5.72,"file":"--cif7Nv6BI_chunk_008_scene_014_choices.mp4"},{"index":8,"timestamp":6.76,"file":"--cif7Nv6BI_chunk_008_scene_014_choices.mp4"},{"index":9,"timestamp":7.8,"file":"--cif7Nv6BI_chunk_008_scene_014_choices.mp4"},{"index":10,"timestamp":8.84,"file":"--cif7Nv6BI_chunk_008_scene_014_choices.mp4"},{"index":11,"timestamp":9.88,"file":"--cif7Nv6BI_chunk_008_scene_014_choices.mp4"},{"index":12,"timestamp":10.66,"file":"--cif7Nv6BI_chunk_008_scene_014_choices.mp4"},{"index":13,"timestamp":11.700000000000001,"file":"--cif7Nv6BI_chunk_008_scene_014_choices.mp4"}]}},"select":{"inputs":{"model":"gemini-3.5-flash-lite","temperature":0.0,"as_video":true,"media_bytes":3587038,"media":["--cif7Nv6BI_chunk_008_scene_014_choices.mp4"],"choice_frames":13,"system_prompt":"You are labelling video data for referring multi-object tracking.\n\nYou will see a numbered sequence of frames from one clip, in time order. Each frame has its own number and timestamp printed in the top-left corner. Candidate objects have been detected for you and drawn on the frames as coloured boxes, each labelled with a number. The same object keeps the same number and the same colour on every frame. Below each frame is a strip of enlarged crops of those same numbered boxes, so you can identify a small or distant object.\n\nFor each description you are given, answer two questions:\n\n1. MATCHING_IDS — which of the numbered objects the description refers to. List every number that matches, or an EMPTY LIST if none of them do. An empty list is a correct and expected answer: some descriptions are deliberately about something that is not in this clip. Do not pick a number just to have picked one, and do not pick a number for an object that merely resembles the description.\n\n2. START_FRAME and END_FRAME — the frame numbers between which the described state actually holds. If the description is \"the car that's turning left\", the interval is the frames in which that car is turning, NOT every frame in which the car is visible. Use the frame numbers printed on the frames. START_FRAME must not be greater than END_FRAME. When MATCHING_IDS is empty, give the full range (the first and last frame numbers); it is ignored.\n\n3. CONFIDENCE — how sure you are of the MATCHING_IDS answer, from 0.0 to 1.0. Be honest: a low number costs the example nothing, and an overconfident wrong answer becomes a wrong label.\n\nRules:\n- Only use numbers that are actually drawn on the frames.\n- Never output coordinates, box values, or pixel positions.\n- Judge motion by comparing the object's position across frames, not by guessing from one frame.\n","user_prompt":"The video has 13 frames, numbered 1 to 13. Each frame is one moment of the clip and has its number burned into the top left corner.\n\nAnswer for each of the following descriptions, in the order given:\n1. the man standing on the stone path\n2. the dark dog running down the hill","questions":["the man standing on the stone path","the dark dog running down the hill"]},"outputs":{"raw_response":"{\n  \"answers\": [\n    {\n      \"description_index\": 1,\n      \"matching_ids\": [1],\n      \"start_frame\": 1,\n      \"end_frame\": 3,\n      \"confidence\": 1.0\n    },\n    {\n      \"description_index\": 2,\n      \"matching_ids\": [],\n      \"start_frame\": 1,\n      \"end_frame\": 13,\n      \"confidence\": 1.0\n    }\n  ]\n}","raw_answers":[{"description_index":1,"matching_ids":[1],"start_frame":1,"end_frame":3,"confidence":1.0},{"description_index":2,"matching_ids":[],"start_frame":1,"end_frame":13,"confidence":1.0}],"answers":[{"expression":"the man standing on the stone path","matching_presentation_ids":[1],"start_choice_frame":1,"end_choice_frame":3,"confidence":1.0,"self_reported_confidence":1.0},{"expression":"the dark dog running down the hill","matching_presentation_ids":[],"start_choice_frame":1,"end_choice_frame":13,"confidence":1.0,"self_reported_confidence":1.0}]}},"verify":{"inputs":{"expressions":2,"tracklets":1,"selections":2,"gates":"interval, count, unknown ids, confidence, then geometry (motion)"},"outputs":{"kept":[{"expression":"the man standing on the stone path","kind":"appearance","is_negative":false,"tracks":[{"timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08],"boxes":[[0.5973958373069763,0.3638888895511627,0.6526041738688946,0.6416666507720947],[0.5838541984558105,0.385185182094574,0.6385417021811008,0.6462962925434113],[0.5614583492279053,0.40740740299224854,0.6062500178813934,0.6611111164093018],[0.5640625357627869,0.4138888716697693,0.6114583723247051,0.6712962687015533],[0.5281250476837158,0.4268518388271332,0.5791667178273201,0.6925925612449646],[0.5083333849906921,0.4120370149612427,0.5744792222976685,0.680555522441864],[0.4932291805744171,0.4203703701496124,0.5473958514630795,0.6972222030162811],[0.4567708671092987,0.4435185194015503,0.511979203671217,0.7287037074565887],[0.3723958432674408,0.4287036955356598,0.42239584773778915,0.7249999940395355]]}]}],"rejected":{"the dark dog running down the hill":"count_mismatch"},"rejection_counts":{"count_mismatch":1},"row":{"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_008_scene_014.mp4","scene_key":"--cif7Nv6BI_chunk_008_scene_014","category":"test_vehicles","height":1080,"width":1920,"fps":50.0,"num_frames":594,"duration":11.88,"rmot_expressions":[{"expression":"the man standing on the stone path","kind":"appearance","is_negative":false,"tracks":[{"timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08],"boxes":[[0.5973958373069763,0.3638888895511627,0.6526041738688946,0.6416666507720947],[0.5838541984558105,0.385185182094574,0.6385417021811008,0.6462962925434113],[0.5614583492279053,0.40740740299224854,0.6062500178813934,0.6611111164093018],[0.5640625357627869,0.4138888716697693,0.6114583723247051,0.6712962687015533],[0.5281250476837158,0.4268518388271332,0.5791667178273201,0.6925925612449646],[0.5083333849906921,0.4120370149612427,0.5744792222976685,0.680555522441864],[0.4932291805744171,0.4203703701496124,0.5473958514630795,0.6972222030162811],[0.4567708671092987,0.4435185194015503,0.511979203671217,0.7287037074565887],[0.3723958432674408,0.4287036955356598,0.42239584773778915,0.7249999940395355]]}]}],"rmot_expression_count":1,"rmot_negative_count":0,"rmot_motion_count":0}}}}},{"index":5,"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_001_scene_010.mp4","status":"complete","steps":{"decode":{"inputs":{"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_001_scene_010.mp4","sample_fps":4.0},"outputs":{"width":1920,"height":1080,"source_fps":50.0,"duration_seconds":14.72,"frames_kept":57,"blank_frames_dropped":0,"timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88,10.14,10.4,10.66,10.92,11.18,11.44,11.700000000000001,11.96,12.22,12.48,12.74,13.0,13.26,13.52,13.780000000000001,14.040000000000001,14.3,14.56],"error":""}},"propose":{"inputs":{"model":"gemini-3.5-flash-lite","temperature":0.4,"video_bytes":14433278,"system_prompt":"You are annotating video clips to build training data for referring multi-object tracking: given a video and a natural-language description, a model must output the boxes of every object matching that description, over time.\n\nYour job is to write the descriptions. You do NOT output coordinates, boxes, pixel positions, or any numbers about where things are. A separate detector finds the objects.\n\nFor every clip you produce a list of items. Each item has two texts that must NOT be the same:\n\nEXPRESSION — what a person would say to single out the object or objects. It may describe motion, temporal order, interaction with other objects, and spatial relations. It should read like something a person would actually say.\n  Good: \"the black car that's turning left\"\n  Good: \"the cyclist who overtakes the bus\"\n  Good: \"the two people who walk out of frame on the right\"\n\nNOUN_PHRASE — the same object as a plain, timeless visual category with its visible attributes, and NOTHING else. The detector is prompted with this text and looks at ONE frame at a time, so anything about time, order, or action is useless to it and makes it fail.\n  Allowed: category nouns, colours, materials, clothing, size, type\n  Forbidden: verbs and participles of action (\"turning\", \"walks\", \"is moving\"), temporal words (\"then\", \"starts\", \"after\", \"first\", \"last\"), relative clauses (\"that ...\", \"which ...\", \"who ...\"), ordinals and rank words (\"first\", \"leftmost\", \"nearest\"), and anything over six words.\n  \"the black car that's turning left\" -> \"black car\"\n  \"the cyclist who overtakes the bus\" -> \"cyclist\"\n  \"the two people who walk out of frame on the right\" -> \"person\"\n\nKIND — what makes the expression hard: \"motion\", \"appearance\", \"spatial\", or \"interaction\". Aim for about 35% \"motion\" items: motion is the part a single-frame model cannot already solve, so it is the most valuable, but a list of only motion items is not usable training data either.\n\nEXPECTED_COUNT — how many distinct objects in the clip match the EXPRESSION. Count carefully; this is checked against the detector's answer, and a wrong count discards the item.\n\nIS_NEGATIVE — at most ONE item may be a negative, and only when you also write at least one non-negative item. A negative is an expression that recombines attributes that ARE present in the clip into a whole that is NOT. If the only black car is moving, \"the parked black car\" is a good negative. Its NOUN_PHRASE must still be the category that IS present (\"black car\"), because that is what makes the absence checkable rather than merely unverified. A negative's EXPECTED_COUNT is 0.\n\nRules that discard an item, so follow them exactly:\n- Never write coordinates, box values, pixel positions, or percentages of the frame.\n- Never let a NOUN_PHRASE contain a verb of action, a temporal word, a relative clause, or an ordinal.\n- Never write an EXPRESSION whose object is not visible in the clip, unless it is marked IS_NEGATIVE.\n- Prefer objects that are clearly visible and identifiable over background clutter.\n- Do not describe the camera or the shot; describe objects in the scene.\n","user_prompt":"Watch this clip and write between 1 and 3 items following the rules. Cover different objects rather than restating one object several ways, and include any negative in the same list.\n"},"outputs":{"raw_response":"{\n  \"items\": [\n    {\n      \"expression\": \"the white rooster weather vane on top of the church roof\",\n      \"noun_phrase\": \"white rooster weather vane\",\n      \"kind\": \"appearance\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the stone church building with an arched wooden door\",\n      \"noun_phrase\": \"stone church building\",\n      \"kind\": \"appearance\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the modern red car parked on the street\",\n      \"noun_phrase\": \"red car\",\n      \"kind\": \"appearance\",\n      \"expected_count\": 0,\n      \"is_negative\": true\n    }\n  ]\n}","raw_items":[{"expression":"the white rooster weather vane on top of the church roof","noun_phrase":"white rooster weather vane","kind":"appearance","expected_count":1,"is_negative":false},{"expression":"the stone church building with an arched wooden door","noun_phrase":"stone church building","kind":"appearance","expected_count":1,"is_negative":false},{"expression":"the modern red car parked on the street","noun_phrase":"red car","kind":"appearance","expected_count":0,"is_negative":true}],"accepted":[{"expression":"the white rooster weather vane on top of the church roof","noun_phrase":"white rooster weather vane","kind":"appearance","expected_count":1,"is_negative":false},{"expression":"the stone church building with an arched wooden door","noun_phrase":"stone church building","kind":"appearance","expected_count":1,"is_negative":false},{"expression":"the modern red car parked on the street","noun_phrase":"red car","kind":"appearance","expected_count":0,"is_negative":true}],"gated":[]}},"track":{"inputs":{"tracker":"Sam31MultiplexTracker","model_id":"/mnt/ceph3/ec/jacky/sam3.1","noun_phrases":["red car","stone church building","white rooster weather vane"],"expressions_in":3,"frames":57,"sample_fps":4.0,"frame_size":"1920x1080","timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88,10.14,10.4,10.66,10.92,11.18,11.44,11.700000000000001,11.96,12.22,12.48,12.74,13.0,13.26,13.52,13.780000000000001,14.040000000000001,14.3,14.56],"session":"one session over 57 JPEG frames; then per phrase: reset_session, add_prompt(text) on frame 0, propagate_in_video","passes":3},"outputs":{"raw_tracks":[{"noun_phrase":"stone church building","object_id":0,"observations":[{"timestamp":0.0,"box":[0.40260419249534607,0.4175925850868225,0.999479204416275,0.970370352268219],"occluded":false},{"timestamp":0.26,"box":[0.3994791805744171,0.4175925850868225,0.999479204416275,0.9712962508201599],"occluded":false},{"timestamp":0.52,"box":[0.3932291865348816,0.4166666567325592,0.9994792342185974,0.9712962806224823],"occluded":false},{"timestamp":0.78,"box":[0.3864583671092987,0.4148148000240326,0.999479204416275,0.9712962806224823],"occluded":false},{"timestamp":1.04,"box":[0.38072919845581055,0.4157407283782959,0.9994792342185974,0.9722222089767456],"occluded":false},{"timestamp":1.3,"box":[0.37291669845581055,0.4138888716697693,0.9994792342185974,0.9722222089767456],"occluded":false},{"timestamp":1.56,"box":[0.3656250238418579,0.4055555462837219,0.9994792342185974,0.9722222089767456],"occluded":false},{"timestamp":1.82,"box":[0.36406251788139343,0.4018518328666687,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":2.08,"box":[0.3723958432674408,0.39537036418914795,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":2.34,"box":[0.3739583492279053,0.39629629254341125,0.9994792342185974,0.9722221791744232],"occluded":false},{"timestamp":2.6,"box":[0.36927086114883423,0.3888888955116272,0.9994792342185974,0.9722222089767456],"occluded":false},{"timestamp":2.86,"box":[0.3661458492279053,0.37962961196899414,0.9994792342185974,0.9722222089767456],"occluded":false},{"timestamp":3.12,"box":[0.37968751788139343,0.3916666507720947,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":3.38,"box":[0.38593751192092896,0.40925925970077515,0.9994792342185974,0.9722222089767456],"occluded":false},{"timestamp":3.64,"box":[0.3817708492279053,0.43611109256744385,0.9994792342185974,0.9712962508201599],"occluded":false},{"timestamp":3.9,"box":[0.3812500238418579,0.4453703761100769,0.9994792342185974,0.9712963104248047],"occluded":false},{"timestamp":4.16,"box":[0.3802083432674408,0.44722220301628113,0.999479204416275,0.9712962806224823],"occluded":false},{"timestamp":4.42,"box":[0.3854166865348816,0.4342592656612396,0.9994792342185974,0.9712962806224823],"occluded":false},{"timestamp":4.68,"box":[0.3921875059604645,0.42222222685813904,0.999479204416275,0.9712962806224823],"occluded":false},{"timestamp":4.94,"box":[0.3958333432674408,0.41018518805503845,0.999479204416275,0.9712962806224823],"occluded":false},{"timestamp":5.2,"box":[0.39635419845581055,0.4037036895751953,0.9994792342185974,0.9722222089767456],"occluded":false},{"timestamp":5.46,"box":[0.3984375298023224,0.4000000059604645,0.999479204416275,0.9712962806224823],"occluded":false},{"timestamp":5.72,"box":[0.40000003576278687,0.39907407760620117,0.9994792342185974,0.9712963104248047],"occluded":false},{"timestamp":5.98,"box":[0.40156251192092896,0.4009259045124054,0.9994792342185974,0.9722221791744232],"occluded":false},{"timestamp":6.24,"box":[0.40312501788139343,0.4009259045124054,0.999479204416275,0.9722221791744232],"occluded":false},{"timestamp":6.5,"box":[0.4036458432674408,0.4018518328666687,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":6.76,"box":[0.40156251192092896,0.4046296179294586,0.9994792342185974,0.9712962806224823],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.40156251192092896,0.4046296179294586,0.9994792342185974,0.9722221791744232],"occluded":false},{"timestamp":7.28,"box":[0.39895835518836975,0.40648147463798523,0.999479204416275,0.9722221791744232],"occluded":false},{"timestamp":7.54,"box":[0.39531251788139343,0.40925925970077515,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":7.8,"box":[0.39375001192092896,0.40925925970077515,0.9994792342185974,0.9712963104248047],"occluded":false},{"timestamp":8.06,"box":[0.3921875059604645,0.4055555462837219,0.999479204416275,0.9712962508201599],"occluded":false},{"timestamp":8.32,"box":[0.3932291865348816,0.39907407760620117,0.9994792342185974,0.9712963104248047],"occluded":false},{"timestamp":8.58,"box":[0.3880208432674408,0.3898147940635681,0.999479204416275,0.9712962508201599],"occluded":false},{"timestamp":8.84,"box":[0.38593751192092896,0.385185182094574,0.9994792342185974,0.9722222089767456],"occluded":false},{"timestamp":9.1,"box":[0.3921875059604645,0.3768518567085266,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":9.36,"box":[0.3942708671092987,0.34351852536201477,0.999479204416275,0.9712962806224823],"occluded":false},{"timestamp":9.620000000000001,"box":[0.41458335518836975,0.3324074149131775,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":9.88,"box":[0.45000001788139343,0.345370352268219,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":10.14,"box":[0.46302086114883423,0.36574074625968933,0.9994792342185974,0.9712962806224823],"occluded":false},{"timestamp":10.4,"box":[0.46302086114883423,0.3777777850627899,0.9994792342185974,0.972222238779068],"occluded":false},{"timestamp":10.66,"box":[0.4697916805744171,0.36759257316589355,0.999479204416275,0.9712962508201599],"occluded":false},{"timestamp":10.92,"box":[0.4895833730697632,0.345370352268219,0.9994792342185974,0.9694443941116333],"occluded":false},{"timestamp":11.18,"box":[0.5062500238418579,0.3305555582046509,0.999479204416275,0.9157407283782959],"occluded":false},{"timestamp":11.44,"box":[0.5182291865348816,0.3305555582046509,0.999479204416275,0.9222221970558167],"occluded":false},{"timestamp":11.700000000000001,"box":[0.5338541865348816,0.32685184478759766,0.999479204416275,0.9666666388511658],"occluded":false},{"timestamp":11.96,"box":[0.5385416746139526,0.27314814925193787,0.999479204416275,0.8805555403232574],"occluded":false},{"timestamp":12.22,"box":[0.5359375476837158,0.24537037312984467,0.9994792342185974,0.8527777642011642],"occluded":false},{"timestamp":12.48,"box":[0.5088542103767395,0.20370370149612427,0.9994792342185974,0.7972221970558167],"occluded":false},{"timestamp":12.74,"box":[0.5229166746139526,0.1666666567325592,0.999479204416275,0.7898148000240326],"occluded":false},{"timestamp":13.0,"box":[0.5317708849906921,0.21574074029922485,0.9994792342185974,0.8527777791023254],"occluded":false},{"timestamp":13.26,"box":[0.5395833849906921,0.18796296417713165,0.9994792342185974,0.8592592626810074],"occluded":false},{"timestamp":13.52,"box":[0.5291666984558105,0.17037037014961243,0.9994792342185974,0.8611110746860504],"occluded":false},{"timestamp":13.780000000000001,"box":[0.5083333849906921,0.18333333730697632,0.9994792342185974,0.8824074268341064],"occluded":false},{"timestamp":14.040000000000001,"box":[0.5223958492279053,0.1509259194135666,0.999479204416275,0.8805555552244186],"occluded":false},{"timestamp":14.3,"box":[0.5479167103767395,0.14537036418914795,0.9994792342185974,0.9138888716697693],"occluded":false},{"timestamp":14.56,"box":[0.47708335518836975,0.13703703880310059,0.999479204416275,0.9361110925674438],"occluded":false}],"scores":[0.7697842121124268,0.7697842121124268,0.7697842121124268,0.7697842121124268,0.7697842121124268,0.7697842121124268,0.7697842121124268,0.7697842121124268,0.7697842121124268,0.7697842121124268,0.7697842121124268,0.7697842121124268,0.7697842121124268,0.7697842121124268,0.7697842121124268,0.7697842121124268,0.7697842121124268,0.7697842121124268,0.7697842121124268,0.7697842121124268,0.7697842121124268,0.7697842121124268,0.7697842121124268,0.7697842121124268,0.7697842121124268,0.7697842121124268,0.7697842121124268,0.7697842121124268,0.7697842121124268,0.7697842121124268,0.7697842121124268,0.7697842121124268,0.7697842121124268,0.7697842121124268,0.7697842121124268,0.7697842121124268,0.7697842121124268,0.7697842121124268,0.7697842121124268,0.7697842121124268,0.7697842121124268,0.7697842121124268,0.7697842121124268,0.7697842121124268,0.7697842121124268,0.7697842121124268,0.7697842121124268,0.7697842121124268,0.7697842121124268,0.7697842121124268,0.7697842121124268,0.7697842121124268,0.7697842121124268,0.7697842121124268,0.7697842121124268,0.7697842121124268,0.7697842121124268]},{"noun_phrase":"white rooster weather vane","object_id":0,"observations":[{"timestamp":0.0,"box":[0.4765625298023224,0.4462962746620178,0.49062503036111593,0.4879629388451576],"occluded":false},{"timestamp":0.26,"box":[0.47864586114883423,0.44907405972480774,0.49427086114883423,0.4953703545033932],"occluded":false},{"timestamp":0.52,"box":[0.47812503576278687,0.4546296298503876,0.633333370089531,0.6435185223817825],"occluded":false},{"timestamp":0.78,"box":[0.47864586114883423,0.4546296298503876,0.49635419622063637,0.5064814798533916],"occluded":false},{"timestamp":1.04,"box":[0.47812503576278687,0.4583333134651184,0.6473958790302277,0.6990740448236465],"occluded":false},{"timestamp":1.3,"box":[0.47447919845581055,0.4592592418193817,0.49270836636424065,0.5166666470468044],"occluded":false},{"timestamp":1.56,"box":[0.47031253576278687,0.45277777314186096,0.4890625365078449,0.511111106723547],"occluded":false},{"timestamp":1.82,"box":[0.47031253576278687,0.44907405972480774,0.4895833693444729,0.5083333179354668],"occluded":false},{"timestamp":2.08,"box":[0.4776041805744171,0.4435185194015503,0.661458358168602,0.6990740597248077],"occluded":false},{"timestamp":2.34,"box":[0.4791666865348816,0.4453703761100769,0.4984375201165676,0.5046296343207359],"occluded":false},{"timestamp":2.6,"box":[0.47447919845581055,0.43796294927597046,0.49375003203749657,0.4962962828576565],"occluded":false},{"timestamp":2.86,"box":[0.47187501192092896,0.4287036955356598,0.491145845502615,0.4888888783752918],"occluded":false},{"timestamp":3.12,"box":[0.483854204416275,0.4416666626930237,0.5036458726972342,0.4999999962747097],"occluded":false},{"timestamp":3.38,"box":[0.4911458492279053,0.4592592418193817,0.6750000268220901,0.7129629552364349],"occluded":false},{"timestamp":3.64,"box":[0.48645836114883423,0.4861111044883728,0.5062500294297934,0.5462962873280048],"occluded":false},{"timestamp":3.9,"box":[0.48645836114883423,0.4962962865829468,0.5057291947305202,0.5564814694225788],"occluded":false},{"timestamp":4.16,"box":[0.48489585518836975,0.4981481432914734,0.5046875234693289,0.5574074015021324],"occluded":false},{"timestamp":4.42,"box":[0.4895833730697632,0.4851851761341095,0.6755208820104599,0.7407407164573669],"occluded":false},{"timestamp":4.68,"box":[0.4958333671092987,0.4731481373310089,0.5151042006909847,0.5324073955416679],"occluded":false},{"timestamp":4.94,"box":[0.499479204416275,0.4611110985279083,0.518750037997961,0.5203703567385674],"occluded":false},{"timestamp":5.2,"box":[0.5005208849906921,0.4546296298503876,0.5203125532716513,0.5166666656732559],"occluded":false},{"timestamp":5.46,"box":[0.5026041865348816,0.45092591643333435,0.6890625357627869,0.6990740597248077],"occluded":false},{"timestamp":5.72,"box":[0.5041667222976685,0.44999998807907104,0.5234375558793545,0.5092592462897301],"occluded":false},{"timestamp":5.98,"box":[0.5062500238418579,0.45185184478759766,0.5255208574235439,0.5120370276272297],"occluded":false},{"timestamp":6.24,"box":[0.5078125,0.45185184478759766,0.527083333581686,0.5120370276272297],"occluded":false},{"timestamp":6.5,"box":[0.5083333849906921,0.45277777314186096,0.5281250532716513,0.51203703135252],"occluded":false},{"timestamp":6.76,"box":[0.5062500238418579,0.45370370149612427,0.6890625357627869,0.7129629552364349],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.5057291984558105,0.4555555582046509,0.5255208667367697,0.5138888917863369],"occluded":false},{"timestamp":7.28,"box":[0.5041667222976685,0.4555555582046509,0.6890625655651093,0.7129629552364349],"occluded":false},{"timestamp":7.54,"box":[0.5010417103767395,0.4564814865589142,0.6890625506639481,0.6990740746259689],"occluded":false},{"timestamp":7.8,"box":[0.499479204416275,0.4574073851108551,0.518750037997961,0.5175925679504871],"occluded":false},{"timestamp":8.06,"box":[0.49791669845581055,0.4546296298503876,0.5177083667367697,0.5129629634320736],"occluded":false},{"timestamp":8.32,"box":[0.4989583492279053,0.44722220301628113,0.5187500175088644,0.5074073858559132],"occluded":false},{"timestamp":8.58,"box":[0.49427086114883423,0.43703702092170715,0.5135416947305202,0.4962962791323662],"occluded":false},{"timestamp":8.84,"box":[0.48906251788139343,0.4305555522441864,0.5078125186264515,0.4888888858258724],"occluded":false},{"timestamp":9.1,"box":[0.491666704416275,0.42222222685813904,0.5098958723247051,0.47777778282761574],"occluded":false},{"timestamp":9.36,"box":[0.4906250238418579,0.3879629671573639,0.6614583730697632,0.6018518507480621],"occluded":false},{"timestamp":9.620000000000001,"box":[0.5062500238418579,0.3731481432914734,0.5234375242143869,0.42407406866550446],"occluded":false},{"timestamp":9.88,"box":[0.534375011920929,0.38240739703178406,0.5505208466202021,0.4287036918103695],"occluded":false},{"timestamp":10.14,"box":[0.5375000238418579,0.4009259045124054,0.55260419100523,0.4435184970498085],"occluded":false},{"timestamp":10.4,"box":[0.5270833373069763,0.41111111640930176,0.5401041712611914,0.45000000298023224],"occluded":false},{"timestamp":10.66,"box":[0.5276042222976685,0.4009259045124054,0.5401042234152555,0.43611108884215355],"occluded":false},{"timestamp":10.92,"box":[0.5385416746139526,0.38055554032325745,0.5500000081956387,0.4148148000240326],"occluded":false},{"timestamp":11.18,"box":[0.550000011920929,0.36851850152015686,0.560937512665987,0.3981481306254864],"occluded":false},{"timestamp":11.44,"box":[0.5609375238418579,0.3722222149372101,0.5713541908189654,0.3990740664303303],"occluded":false},{"timestamp":11.700000000000001,"box":[0.5750000476837158,0.3722222149372101,0.5854167146608233,0.3972222153097391],"occluded":false},{"timestamp":11.96,"box":[0.5791667103767395,0.32685184478759766,0.5901042111217976,0.3462962880730629],"occluded":false},{"timestamp":12.22,"box":[0.5765625238418579,0.3064814805984497,0.5848958576098084,0.3203703695908189],"occluded":false},{"timestamp":12.48,"box":[0.5479167103767395,0.26851850748062134,0.5567708779126406,0.2824073964729905],"occluded":false}],"scores":[0.8476190567016602,0.8476190567016602,0.8476190567016602,0.8476190567016602,0.8476190567016602,0.8476190567016602,0.8476190567016602,0.8476190567016602,0.8476190567016602,0.8476190567016602,0.8476190567016602,0.8476190567016602,0.8476190567016602,0.8476190567016602,0.8476190567016602,0.8476190567016602,0.8476190567016602,0.8476190567016602,0.8476190567016602,0.8476190567016602,0.8476190567016602,0.8476190567016602,0.8476190567016602,0.8476190567016602,0.8476190567016602,0.8476190567016602,0.8476190567016602,0.8476190567016602,0.8476190567016602,0.8476190567016602,0.8476190567016602,0.8476190567016602,0.8476190567016602,0.8476190567016602,0.8476190567016602,0.8476190567016602,0.8476190567016602,0.8476190567016602,0.8476190567016602,0.8476190567016602,0.8476190567016602,0.8476190567016602,0.8476190567016602,0.8476190567016602,0.8476190567016602,0.8476190567016602,0.8476190567016602,0.8476190567016602,0.8476190567016602]}],"tracklets":[{"noun_phrase":"white rooster weather vane","object_id":0,"presentation_id":1,"observations":[{"timestamp":0.0,"box":[0.4765625298023224,0.4462962746620178,0.49062503036111593,0.4879629388451576],"occluded":false},{"timestamp":0.26,"box":[0.47864586114883423,0.44907405972480774,0.49427086114883423,0.4953703545033932],"occluded":false},{"timestamp":0.52,"box":[0.47812503576278687,0.4546296298503876,0.633333370089531,0.6435185223817825],"occluded":false},{"timestamp":0.78,"box":[0.47864586114883423,0.4546296298503876,0.49635419622063637,0.5064814798533916],"occluded":false},{"timestamp":1.04,"box":[0.47812503576278687,0.4583333134651184,0.6473958790302277,0.6990740448236465],"occluded":false},{"timestamp":1.3,"box":[0.47447919845581055,0.4592592418193817,0.49270836636424065,0.5166666470468044],"occluded":false},{"timestamp":1.56,"box":[0.47031253576278687,0.45277777314186096,0.4890625365078449,0.511111106723547],"occluded":false},{"timestamp":1.82,"box":[0.47031253576278687,0.44907405972480774,0.4895833693444729,0.5083333179354668],"occluded":false},{"timestamp":2.08,"box":[0.4776041805744171,0.4435185194015503,0.661458358168602,0.6990740597248077],"occluded":false},{"timestamp":2.34,"box":[0.4791666865348816,0.4453703761100769,0.4984375201165676,0.5046296343207359],"occluded":false},{"timestamp":2.6,"box":[0.47447919845581055,0.43796294927597046,0.49375003203749657,0.4962962828576565],"occluded":false},{"timestamp":2.86,"box":[0.47187501192092896,0.4287036955356598,0.491145845502615,0.4888888783752918],"occluded":false},{"timestamp":3.12,"box":[0.483854204416275,0.4416666626930237,0.5036458726972342,0.4999999962747097],"occluded":false},{"timestamp":3.38,"box":[0.4911458492279053,0.4592592418193817,0.6750000268220901,0.7129629552364349],"occluded":false},{"timestamp":3.64,"box":[0.48645836114883423,0.4861111044883728,0.5062500294297934,0.5462962873280048],"occluded":false},{"timestamp":3.9,"box":[0.48645836114883423,0.4962962865829468,0.5057291947305202,0.5564814694225788],"occluded":false},{"timestamp":4.16,"box":[0.48489585518836975,0.4981481432914734,0.5046875234693289,0.5574074015021324],"occluded":false},{"timestamp":4.42,"box":[0.4895833730697632,0.4851851761341095,0.6755208820104599,0.7407407164573669],"occluded":false},{"timestamp":4.68,"box":[0.4958333671092987,0.4731481373310089,0.5151042006909847,0.5324073955416679],"occluded":false},{"timestamp":4.94,"box":[0.499479204416275,0.4611110985279083,0.518750037997961,0.5203703567385674],"occluded":false},{"timestamp":5.2,"box":[0.5005208849906921,0.4546296298503876,0.5203125532716513,0.5166666656732559],"occluded":false},{"timestamp":5.46,"box":[0.5026041865348816,0.45092591643333435,0.6890625357627869,0.6990740597248077],"occluded":false},{"timestamp":5.72,"box":[0.5041667222976685,0.44999998807907104,0.5234375558793545,0.5092592462897301],"occluded":false},{"timestamp":5.98,"box":[0.5062500238418579,0.45185184478759766,0.5255208574235439,0.5120370276272297],"occluded":false},{"timestamp":6.24,"box":[0.5078125,0.45185184478759766,0.527083333581686,0.5120370276272297],"occluded":false},{"timestamp":6.5,"box":[0.5083333849906921,0.45277777314186096,0.5281250532716513,0.51203703135252],"occluded":false},{"timestamp":6.76,"box":[0.5062500238418579,0.45370370149612427,0.6890625357627869,0.7129629552364349],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.5057291984558105,0.4555555582046509,0.5255208667367697,0.5138888917863369],"occluded":false},{"timestamp":7.28,"box":[0.5041667222976685,0.4555555582046509,0.6890625655651093,0.7129629552364349],"occluded":false},{"timestamp":7.54,"box":[0.5010417103767395,0.4564814865589142,0.6890625506639481,0.6990740746259689],"occluded":false},{"timestamp":7.8,"box":[0.499479204416275,0.4574073851108551,0.518750037997961,0.5175925679504871],"occluded":false},{"timestamp":8.06,"box":[0.49791669845581055,0.4546296298503876,0.5177083667367697,0.5129629634320736],"occluded":false},{"timestamp":8.32,"box":[0.4989583492279053,0.44722220301628113,0.5187500175088644,0.5074073858559132],"occluded":false},{"timestamp":8.58,"box":[0.49427086114883423,0.43703702092170715,0.5135416947305202,0.4962962791323662],"occluded":false},{"timestamp":8.84,"box":[0.48906251788139343,0.4305555522441864,0.5078125186264515,0.4888888858258724],"occluded":false},{"timestamp":9.1,"box":[0.491666704416275,0.42222222685813904,0.5098958723247051,0.47777778282761574],"occluded":false},{"timestamp":9.36,"box":[0.4906250238418579,0.3879629671573639,0.6614583730697632,0.6018518507480621],"occluded":false},{"timestamp":9.620000000000001,"box":[0.5062500238418579,0.3731481432914734,0.5234375242143869,0.42407406866550446],"occluded":false},{"timestamp":9.88,"box":[0.534375011920929,0.38240739703178406,0.5505208466202021,0.4287036918103695],"occluded":false},{"timestamp":10.14,"box":[0.5375000238418579,0.4009259045124054,0.55260419100523,0.4435184970498085],"occluded":false},{"timestamp":10.4,"box":[0.5270833373069763,0.41111111640930176,0.5401041712611914,0.45000000298023224],"occluded":false},{"timestamp":10.66,"box":[0.5276042222976685,0.4009259045124054,0.5401042234152555,0.43611108884215355],"occluded":false},{"timestamp":10.92,"box":[0.5385416746139526,0.38055554032325745,0.5500000081956387,0.4148148000240326],"occluded":false},{"timestamp":11.18,"box":[0.550000011920929,0.36851850152015686,0.560937512665987,0.3981481306254864],"occluded":false},{"timestamp":11.44,"box":[0.5609375238418579,0.3722222149372101,0.5713541908189654,0.3990740664303303],"occluded":false},{"timestamp":11.700000000000001,"box":[0.5750000476837158,0.3722222149372101,0.5854167146608233,0.3972222153097391],"occluded":false},{"timestamp":11.96,"box":[0.5791667103767395,0.32685184478759766,0.5901042111217976,0.3462962880730629],"occluded":false},{"timestamp":12.22,"box":[0.5765625238418579,0.3064814805984497,0.5848958576098084,0.3203703695908189],"occluded":false},{"timestamp":12.48,"box":[0.5479167103767395,0.26851850748062134,0.5567708779126406,0.2824073964729905],"occluded":false}],"mean_score":0.8476190567016602},{"noun_phrase":"stone church building","object_id":0,"presentation_id":2,"observations":[{"timestamp":0.0,"box":[0.40260419249534607,0.4175925850868225,0.999479204416275,0.970370352268219],"occluded":false},{"timestamp":0.26,"box":[0.3994791805744171,0.4175925850868225,0.999479204416275,0.9712962508201599],"occluded":false},{"timestamp":0.52,"box":[0.3932291865348816,0.4166666567325592,0.9994792342185974,0.9712962806224823],"occluded":false},{"timestamp":0.78,"box":[0.3864583671092987,0.4148148000240326,0.999479204416275,0.9712962806224823],"occluded":false},{"timestamp":1.04,"box":[0.38072919845581055,0.4157407283782959,0.9994792342185974,0.9722222089767456],"occluded":false},{"timestamp":1.3,"box":[0.37291669845581055,0.4138888716697693,0.9994792342185974,0.9722222089767456],"occluded":false},{"timestamp":1.56,"box":[0.3656250238418579,0.4055555462837219,0.9994792342185974,0.9722222089767456],"occluded":false},{"timestamp":1.82,"box":[0.36406251788139343,0.4018518328666687,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":2.08,"box":[0.3723958432674408,0.39537036418914795,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":2.34,"box":[0.3739583492279053,0.39629629254341125,0.9994792342185974,0.9722221791744232],"occluded":false},{"timestamp":2.6,"box":[0.36927086114883423,0.3888888955116272,0.9994792342185974,0.9722222089767456],"occluded":false},{"timestamp":2.86,"box":[0.3661458492279053,0.37962961196899414,0.9994792342185974,0.9722222089767456],"occluded":false},{"timestamp":3.12,"box":[0.37968751788139343,0.3916666507720947,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":3.38,"box":[0.38593751192092896,0.40925925970077515,0.9994792342185974,0.9722222089767456],"occluded":false},{"timestamp":3.64,"box":[0.3817708492279053,0.43611109256744385,0.9994792342185974,0.9712962508201599],"occluded":false},{"timestamp":3.9,"box":[0.3812500238418579,0.4453703761100769,0.9994792342185974,0.9712963104248047],"occluded":false},{"timestamp":4.16,"box":[0.3802083432674408,0.44722220301628113,0.999479204416275,0.9712962806224823],"occluded":false},{"timestamp":4.42,"box":[0.3854166865348816,0.4342592656612396,0.9994792342185974,0.9712962806224823],"occluded":false},{"timestamp":4.68,"box":[0.3921875059604645,0.42222222685813904,0.999479204416275,0.9712962806224823],"occluded":false},{"timestamp":4.94,"box":[0.3958333432674408,0.41018518805503845,0.999479204416275,0.9712962806224823],"occluded":false},{"timestamp":5.2,"box":[0.39635419845581055,0.4037036895751953,0.9994792342185974,0.9722222089767456],"occluded":false},{"timestamp":5.46,"box":[0.3984375298023224,0.4000000059604645,0.999479204416275,0.9712962806224823],"occluded":false},{"timestamp":5.72,"box":[0.40000003576278687,0.39907407760620117,0.9994792342185974,0.9712963104248047],"occluded":false},{"timestamp":5.98,"box":[0.40156251192092896,0.4009259045124054,0.9994792342185974,0.9722221791744232],"occluded":false},{"timestamp":6.24,"box":[0.40312501788139343,0.4009259045124054,0.999479204416275,0.9722221791744232],"occluded":false},{"timestamp":6.5,"box":[0.4036458432674408,0.4018518328666687,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":6.76,"box":[0.40156251192092896,0.4046296179294586,0.9994792342185974,0.9712962806224823],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.40156251192092896,0.4046296179294586,0.9994792342185974,0.9722221791744232],"occluded":false},{"timestamp":7.28,"box":[0.39895835518836975,0.40648147463798523,0.999479204416275,0.9722221791744232],"occluded":false},{"timestamp":7.54,"box":[0.39531251788139343,0.40925925970077515,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":7.8,"box":[0.39375001192092896,0.40925925970077515,0.9994792342185974,0.9712963104248047],"occluded":false},{"timestamp":8.06,"box":[0.3921875059604645,0.4055555462837219,0.999479204416275,0.9712962508201599],"occluded":false},{"timestamp":8.32,"box":[0.3932291865348816,0.39907407760620117,0.9994792342185974,0.9712963104248047],"occluded":false},{"timestamp":8.58,"box":[0.3880208432674408,0.3898147940635681,0.999479204416275,0.9712962508201599],"occluded":false},{"timestamp":8.84,"box":[0.38593751192092896,0.385185182094574,0.9994792342185974,0.9722222089767456],"occluded":false},{"timestamp":9.1,"box":[0.3921875059604645,0.3768518567085266,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":9.36,"box":[0.3942708671092987,0.34351852536201477,0.999479204416275,0.9712962806224823],"occluded":false},{"timestamp":9.620000000000001,"box":[0.41458335518836975,0.3324074149131775,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":9.88,"box":[0.45000001788139343,0.345370352268219,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":10.14,"box":[0.46302086114883423,0.36574074625968933,0.9994792342185974,0.9712962806224823],"occluded":false},{"timestamp":10.4,"box":[0.46302086114883423,0.3777777850627899,0.9994792342185974,0.972222238779068],"occluded":false},{"timestamp":10.66,"box":[0.4697916805744171,0.36759257316589355,0.999479204416275,0.9712962508201599],"occluded":false},{"timestamp":10.92,"box":[0.4895833730697632,0.345370352268219,0.9994792342185974,0.9694443941116333],"occluded":false},{"timestamp":11.18,"box":[0.5062500238418579,0.3305555582046509,0.999479204416275,0.9157407283782959],"occluded":false},{"timestamp":11.44,"box":[0.5182291865348816,0.3305555582046509,0.999479204416275,0.9222221970558167],"occluded":false},{"timestamp":11.700000000000001,"box":[0.5338541865348816,0.32685184478759766,0.999479204416275,0.9666666388511658],"occluded":false},{"timestamp":11.96,"box":[0.5385416746139526,0.27314814925193787,0.999479204416275,0.8805555403232574],"occluded":false},{"timestamp":12.22,"box":[0.5359375476837158,0.24537037312984467,0.9994792342185974,0.8527777642011642],"occluded":false},{"timestamp":12.48,"box":[0.5088542103767395,0.20370370149612427,0.9994792342185974,0.7972221970558167],"occluded":false},{"timestamp":12.74,"box":[0.5229166746139526,0.1666666567325592,0.999479204416275,0.7898148000240326],"occluded":false},{"timestamp":13.0,"box":[0.5317708849906921,0.21574074029922485,0.9994792342185974,0.8527777791023254],"occluded":false},{"timestamp":13.26,"box":[0.5395833849906921,0.18796296417713165,0.9994792342185974,0.8592592626810074],"occluded":false},{"timestamp":13.52,"box":[0.5291666984558105,0.17037037014961243,0.9994792342185974,0.8611110746860504],"occluded":false},{"timestamp":13.780000000000001,"box":[0.5083333849906921,0.18333333730697632,0.9994792342185974,0.8824074268341064],"occluded":false},{"timestamp":14.040000000000001,"box":[0.5223958492279053,0.1509259194135666,0.999479204416275,0.8805555552244186],"occluded":false},{"timestamp":14.3,"box":[0.5479167103767395,0.14537036418914795,0.9994792342185974,0.9138888716697693],"occluded":false},{"timestamp":14.56,"box":[0.47708335518836975,0.13703703880310059,0.999479204416275,0.9361110925674438],"occluded":false}],"mean_score":0.7697842121124268}],"dropped":{},"fates":[{"noun_phrase":"stone church building","object_id":0,"frames":57,"mean_score":0.7697842121124268,"mean_area":0.33708968155877134,"outcome":"kept","presentation_id":2,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"white rooster weather vane","object_id":0,"frames":49,"mean_score":0.8476190567016602,"mean_area":0.009575794754348797,"outcome":"kept","presentation_id":1,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null}],"gates":{"min_track_frames":3,"min_mean_score":0.5,"merge_iou":0.8,"max_tracklets":12}}},"render":{"inputs":{"tracklets":2,"as_video":true,"fps":1.0,"max_frames":30},"outputs":{"media":"/home/zzk/data/video_bbox/vlm_gemini3.5_lite/renders/--cif7Nv6BI_chunk_001_scene_010_choices.mp4","media_bytes":4392051,"frames":[{"index":1,"timestamp":0.0,"file":"--cif7Nv6BI_chunk_001_scene_010_choices.mp4"},{"index":2,"timestamp":1.04,"file":"--cif7Nv6BI_chunk_001_scene_010_choices.mp4"},{"index":3,"timestamp":1.82,"file":"--cif7Nv6BI_chunk_001_scene_010_choices.mp4"},{"index":4,"timestamp":2.86,"file":"--cif7Nv6BI_chunk_001_scene_010_choices.mp4"},{"index":5,"timestamp":3.9,"file":"--cif7Nv6BI_chunk_001_scene_010_choices.mp4"},{"index":6,"timestamp":4.94,"file":"--cif7Nv6BI_chunk_001_scene_010_choices.mp4"},{"index":7,"timestamp":5.72,"file":"--cif7Nv6BI_chunk_001_scene_010_choices.mp4"},{"index":8,"timestamp":6.76,"file":"--cif7Nv6BI_chunk_001_scene_010_choices.mp4"},{"index":9,"timestamp":7.8,"file":"--cif7Nv6BI_chunk_001_scene_010_choices.mp4"},{"index":10,"timestamp":8.84,"file":"--cif7Nv6BI_chunk_001_scene_010_choices.mp4"},{"index":11,"timestamp":9.620000000000001,"file":"--cif7Nv6BI_chunk_001_scene_010_choices.mp4"},{"index":12,"timestamp":10.66,"file":"--cif7Nv6BI_chunk_001_scene_010_choices.mp4"},{"index":13,"timestamp":11.700000000000001,"file":"--cif7Nv6BI_chunk_001_scene_010_choices.mp4"},{"index":14,"timestamp":12.74,"file":"--cif7Nv6BI_chunk_001_scene_010_choices.mp4"},{"index":15,"timestamp":13.52,"file":"--cif7Nv6BI_chunk_001_scene_010_choices.mp4"},{"index":16,"timestamp":14.56,"file":"--cif7Nv6BI_chunk_001_scene_010_choices.mp4"}]}},"select":{"inputs":{"model":"gemini-3.5-flash-lite","temperature":0.0,"as_video":true,"media_bytes":4392051,"media":["--cif7Nv6BI_chunk_001_scene_010_choices.mp4"],"choice_frames":16,"system_prompt":"You are labelling video data for referring multi-object tracking.\n\nYou will see a numbered sequence of frames from one clip, in time order. Each frame has its own number and timestamp printed in the top-left corner. Candidate objects have been detected for you and drawn on the frames as coloured boxes, each labelled with a number. The same object keeps the same number and the same colour on every frame. Below each frame is a strip of enlarged crops of those same numbered boxes, so you can identify a small or distant object.\n\nFor each description you are given, answer two questions:\n\n1. MATCHING_IDS — which of the numbered objects the description refers to. List every number that matches, or an EMPTY LIST if none of them do. An empty list is a correct and expected answer: some descriptions are deliberately about something that is not in this clip. Do not pick a number just to have picked one, and do not pick a number for an object that merely resembles the description.\n\n2. START_FRAME and END_FRAME — the frame numbers between which the described state actually holds. If the description is \"the car that's turning left\", the interval is the frames in which that car is turning, NOT every frame in which the car is visible. Use the frame numbers printed on the frames. START_FRAME must not be greater than END_FRAME. When MATCHING_IDS is empty, give the full range (the first and last frame numbers); it is ignored.\n\n3. CONFIDENCE — how sure you are of the MATCHING_IDS answer, from 0.0 to 1.0. Be honest: a low number costs the example nothing, and an overconfident wrong answer becomes a wrong label.\n\nRules:\n- Only use numbers that are actually drawn on the frames.\n- Never output coordinates, box values, or pixel positions.\n- Judge motion by comparing the object's position across frames, not by guessing from one frame.\n","user_prompt":"The video has 16 frames, numbered 1 to 16. Each frame is one moment of the clip and has its number burned into the top left corner.\n\nAnswer for each of the following descriptions, in the order given:\n1. the white rooster weather vane on top of the church roof\n2. the stone church building with an arched wooden door\n3. the modern red car parked on the street","questions":["the white rooster weather vane on top of the church roof","the stone church building with an arched wooden door","the modern red car parked on the street"]},"outputs":{"raw_response":"{\n  \"answers\": [\n    {\n      \"description_index\": 1,\n      \"matching_ids\": [\n        1\n      ],\n      \"start_frame\": 1,\n      \"end_frame\": 16,\n      \"confidence\": 1.0\n    },\n    {\n      \"description_index\": 2,\n      \"matching_ids\": [\n        2\n      ],\n      \"start_frame\": 1,\n      \"end_frame\": 16,\n      \"confidence\": 1.0\n    },\n    {\n      \"description_index\": 3,\n      \"matching_ids\": [],\n      \"start_frame\": 1,\n      \"end_frame\": 16,\n      \"confidence\": 1.0\n    }\n  ]\n}","raw_answers":[{"description_index":1,"matching_ids":[1],"start_frame":1,"end_frame":16,"confidence":1.0},{"description_index":2,"matching_ids":[2],"start_frame":1,"end_frame":16,"confidence":1.0},{"description_index":3,"matching_ids":[],"start_frame":1,"end_frame":16,"confidence":1.0}],"answers":[{"expression":"the white rooster weather vane on top of the church roof","matching_presentation_ids":[1],"start_choice_frame":1,"end_choice_frame":16,"confidence":1.0,"self_reported_confidence":1.0},{"expression":"the stone church building with an arched wooden door","matching_presentation_ids":[2],"start_choice_frame":1,"end_choice_frame":16,"confidence":1.0,"self_reported_confidence":1.0},{"expression":"the modern red car parked on the street","matching_presentation_ids":[],"start_choice_frame":1,"end_choice_frame":16,"confidence":1.0,"self_reported_confidence":1.0}]}},"verify":{"inputs":{"expressions":3,"tracklets":2,"selections":3,"gates":"interval, count, unknown ids, confidence, then geometry (motion)"},"outputs":{"kept":[{"expression":"the stone church building with an arched wooden door","kind":"appearance","is_negative":false,"tracks":[{"timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88,10.14,10.4,10.66,10.92,11.18,11.44,11.700000000000001,11.96,12.22,12.48,12.74,13.0,13.26,13.52,13.780000000000001,14.040000000000001,14.3,14.56],"boxes":[[0.40260419249534607,0.4175925850868225,0.999479204416275,0.970370352268219],[0.3994791805744171,0.4175925850868225,0.999479204416275,0.9712962508201599],[0.3932291865348816,0.4166666567325592,0.9994792342185974,0.9712962806224823],[0.3864583671092987,0.4148148000240326,0.999479204416275,0.9712962806224823],[0.38072919845581055,0.4157407283782959,0.9994792342185974,0.9722222089767456],[0.37291669845581055,0.4138888716697693,0.9994792342185974,0.9722222089767456],[0.3656250238418579,0.4055555462837219,0.9994792342185974,0.9722222089767456],[0.36406251788139343,0.4018518328666687,0.999479204416275,0.9722222089767456],[0.3723958432674408,0.39537036418914795,0.999479204416275,0.9722222089767456],[0.3739583492279053,0.39629629254341125,0.9994792342185974,0.9722221791744232],[0.36927086114883423,0.3888888955116272,0.9994792342185974,0.9722222089767456],[0.3661458492279053,0.37962961196899414,0.9994792342185974,0.9722222089767456],[0.37968751788139343,0.3916666507720947,0.999479204416275,0.9722222089767456],[0.38593751192092896,0.40925925970077515,0.9994792342185974,0.9722222089767456],[0.3817708492279053,0.43611109256744385,0.9994792342185974,0.9712962508201599],[0.3812500238418579,0.4453703761100769,0.9994792342185974,0.9712963104248047],[0.3802083432674408,0.44722220301628113,0.999479204416275,0.9712962806224823],[0.3854166865348816,0.4342592656612396,0.9994792342185974,0.9712962806224823],[0.3921875059604645,0.42222222685813904,0.999479204416275,0.9712962806224823],[0.3958333432674408,0.41018518805503845,0.999479204416275,0.9712962806224823],[0.39635419845581055,0.4037036895751953,0.9994792342185974,0.9722222089767456],[0.3984375298023224,0.4000000059604645,0.999479204416275,0.9712962806224823],[0.40000003576278687,0.39907407760620117,0.9994792342185974,0.9712963104248047],[0.40156251192092896,0.4009259045124054,0.9994792342185974,0.9722221791744232],[0.40312501788139343,0.4009259045124054,0.999479204416275,0.9722221791744232],[0.4036458432674408,0.4018518328666687,0.999479204416275,0.9722222089767456],[0.40156251192092896,0.4046296179294586,0.9994792342185974,0.9712962806224823],[0.40156251192092896,0.4046296179294586,0.9994792342185974,0.9722221791744232],[0.39895835518836975,0.40648147463798523,0.999479204416275,0.9722221791744232],[0.39531251788139343,0.40925925970077515,0.999479204416275,0.9722222089767456],[0.39375001192092896,0.40925925970077515,0.9994792342185974,0.9712963104248047],[0.3921875059604645,0.4055555462837219,0.999479204416275,0.9712962508201599],[0.3932291865348816,0.39907407760620117,0.9994792342185974,0.9712963104248047],[0.3880208432674408,0.3898147940635681,0.999479204416275,0.9712962508201599],[0.38593751192092896,0.385185182094574,0.9994792342185974,0.9722222089767456],[0.3921875059604645,0.3768518567085266,0.999479204416275,0.9722222089767456],[0.3942708671092987,0.34351852536201477,0.999479204416275,0.9712962806224823],[0.41458335518836975,0.3324074149131775,0.999479204416275,0.9722222089767456],[0.45000001788139343,0.345370352268219,0.999479204416275,0.9722222089767456],[0.46302086114883423,0.36574074625968933,0.9994792342185974,0.9712962806224823],[0.46302086114883423,0.3777777850627899,0.9994792342185974,0.972222238779068],[0.4697916805744171,0.36759257316589355,0.999479204416275,0.9712962508201599],[0.4895833730697632,0.345370352268219,0.9994792342185974,0.9694443941116333],[0.5062500238418579,0.3305555582046509,0.999479204416275,0.9157407283782959],[0.5182291865348816,0.3305555582046509,0.999479204416275,0.9222221970558167],[0.5338541865348816,0.32685184478759766,0.999479204416275,0.9666666388511658],[0.5385416746139526,0.27314814925193787,0.999479204416275,0.8805555403232574],[0.5359375476837158,0.24537037312984467,0.9994792342185974,0.8527777642011642],[0.5088542103767395,0.20370370149612427,0.9994792342185974,0.7972221970558167],[0.5229166746139526,0.1666666567325592,0.999479204416275,0.7898148000240326],[0.5317708849906921,0.21574074029922485,0.9994792342185974,0.8527777791023254],[0.5395833849906921,0.18796296417713165,0.9994792342185974,0.8592592626810074],[0.5291666984558105,0.17037037014961243,0.9994792342185974,0.8611110746860504],[0.5083333849906921,0.18333333730697632,0.9994792342185974,0.8824074268341064],[0.5223958492279053,0.1509259194135666,0.999479204416275,0.8805555552244186],[0.5479167103767395,0.14537036418914795,0.9994792342185974,0.9138888716697693],[0.47708335518836975,0.13703703880310059,0.999479204416275,0.9361110925674438]]}]}],"rejected":{"the modern red car parked on the street":"no_candidates_retrieved","the white rooster weather vane on top of the church roof":"implausible_box"},"rejection_counts":{"no_candidates_retrieved":1,"implausible_box":1},"row":{"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_001_scene_010.mp4","scene_key":"--cif7Nv6BI_chunk_001_scene_010","category":"test_vehicles","height":1080,"width":1920,"fps":50.0,"num_frames":736,"duration":14.72,"rmot_expressions":[{"expression":"the stone church building with an arched wooden door","kind":"appearance","is_negative":false,"tracks":[{"timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88,10.14,10.4,10.66,10.92,11.18,11.44,11.700000000000001,11.96,12.22,12.48,12.74,13.0,13.26,13.52,13.780000000000001,14.040000000000001,14.3,14.56],"boxes":[[0.40260419249534607,0.4175925850868225,0.999479204416275,0.970370352268219],[0.3994791805744171,0.4175925850868225,0.999479204416275,0.9712962508201599],[0.3932291865348816,0.4166666567325592,0.9994792342185974,0.9712962806224823],[0.3864583671092987,0.4148148000240326,0.999479204416275,0.9712962806224823],[0.38072919845581055,0.4157407283782959,0.9994792342185974,0.9722222089767456],[0.37291669845581055,0.4138888716697693,0.9994792342185974,0.9722222089767456],[0.3656250238418579,0.4055555462837219,0.9994792342185974,0.9722222089767456],[0.36406251788139343,0.4018518328666687,0.999479204416275,0.9722222089767456],[0.3723958432674408,0.39537036418914795,0.999479204416275,0.9722222089767456],[0.3739583492279053,0.39629629254341125,0.9994792342185974,0.9722221791744232],[0.36927086114883423,0.3888888955116272,0.9994792342185974,0.9722222089767456],[0.3661458492279053,0.37962961196899414,0.9994792342185974,0.9722222089767456],[0.37968751788139343,0.3916666507720947,0.999479204416275,0.9722222089767456],[0.38593751192092896,0.40925925970077515,0.9994792342185974,0.9722222089767456],[0.3817708492279053,0.43611109256744385,0.9994792342185974,0.9712962508201599],[0.3812500238418579,0.4453703761100769,0.9994792342185974,0.9712963104248047],[0.3802083432674408,0.44722220301628113,0.999479204416275,0.9712962806224823],[0.3854166865348816,0.4342592656612396,0.9994792342185974,0.9712962806224823],[0.3921875059604645,0.42222222685813904,0.999479204416275,0.9712962806224823],[0.3958333432674408,0.41018518805503845,0.999479204416275,0.9712962806224823],[0.39635419845581055,0.4037036895751953,0.9994792342185974,0.9722222089767456],[0.3984375298023224,0.4000000059604645,0.999479204416275,0.9712962806224823],[0.40000003576278687,0.39907407760620117,0.9994792342185974,0.9712963104248047],[0.40156251192092896,0.4009259045124054,0.9994792342185974,0.9722221791744232],[0.40312501788139343,0.4009259045124054,0.999479204416275,0.9722221791744232],[0.4036458432674408,0.4018518328666687,0.999479204416275,0.9722222089767456],[0.40156251192092896,0.4046296179294586,0.9994792342185974,0.9712962806224823],[0.40156251192092896,0.4046296179294586,0.9994792342185974,0.9722221791744232],[0.39895835518836975,0.40648147463798523,0.999479204416275,0.9722221791744232],[0.39531251788139343,0.40925925970077515,0.999479204416275,0.9722222089767456],[0.39375001192092896,0.40925925970077515,0.9994792342185974,0.9712963104248047],[0.3921875059604645,0.4055555462837219,0.999479204416275,0.9712962508201599],[0.3932291865348816,0.39907407760620117,0.9994792342185974,0.9712963104248047],[0.3880208432674408,0.3898147940635681,0.999479204416275,0.9712962508201599],[0.38593751192092896,0.385185182094574,0.9994792342185974,0.9722222089767456],[0.3921875059604645,0.3768518567085266,0.999479204416275,0.9722222089767456],[0.3942708671092987,0.34351852536201477,0.999479204416275,0.9712962806224823],[0.41458335518836975,0.3324074149131775,0.999479204416275,0.9722222089767456],[0.45000001788139343,0.345370352268219,0.999479204416275,0.9722222089767456],[0.46302086114883423,0.36574074625968933,0.9994792342185974,0.9712962806224823],[0.46302086114883423,0.3777777850627899,0.9994792342185974,0.972222238779068],[0.4697916805744171,0.36759257316589355,0.999479204416275,0.9712962508201599],[0.4895833730697632,0.345370352268219,0.9994792342185974,0.9694443941116333],[0.5062500238418579,0.3305555582046509,0.999479204416275,0.9157407283782959],[0.5182291865348816,0.3305555582046509,0.999479204416275,0.9222221970558167],[0.5338541865348816,0.32685184478759766,0.999479204416275,0.9666666388511658],[0.5385416746139526,0.27314814925193787,0.999479204416275,0.8805555403232574],[0.5359375476837158,0.24537037312984467,0.9994792342185974,0.8527777642011642],[0.5088542103767395,0.20370370149612427,0.9994792342185974,0.7972221970558167],[0.5229166746139526,0.1666666567325592,0.999479204416275,0.7898148000240326],[0.5317708849906921,0.21574074029922485,0.9994792342185974,0.8527777791023254],[0.5395833849906921,0.18796296417713165,0.9994792342185974,0.8592592626810074],[0.5291666984558105,0.17037037014961243,0.9994792342185974,0.8611110746860504],[0.5083333849906921,0.18333333730697632,0.9994792342185974,0.8824074268341064],[0.5223958492279053,0.1509259194135666,0.999479204416275,0.8805555552244186],[0.5479167103767395,0.14537036418914795,0.9994792342185974,0.9138888716697693],[0.47708335518836975,0.13703703880310059,0.999479204416275,0.9361110925674438]]}]}],"rmot_expression_count":1,"rmot_negative_count":0,"rmot_motion_count":0}}}}},{"index":6,"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_006_scene_028.mp4","status":"complete_no_labels","steps":{"decode":{"inputs":{"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_006_scene_028.mp4","sample_fps":4.0},"outputs":{"width":1920,"height":1080,"source_fps":50.0,"duration_seconds":16.14,"frames_kept":63,"blank_frames_dropped":0,"timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88,10.14,10.4,10.66,10.92,11.18,11.44,11.700000000000001,11.96,12.22,12.48,12.74,13.0,13.26,13.52,13.780000000000001,14.040000000000001,14.3,14.56,14.82,15.08,15.34,15.6,15.860000000000001,16.12],"error":""}},"propose":{"inputs":{"model":"gemini-3.5-flash-lite","temperature":0.4,"video_bytes":16693150,"system_prompt":"You are annotating video clips to build training data for referring multi-object tracking: given a video and a natural-language description, a model must output the boxes of every object matching that description, over time.\n\nYour job is to write the descriptions. You do NOT output coordinates, boxes, pixel positions, or any numbers about where things are. A separate detector finds the objects.\n\nFor every clip you produce a list of items. Each item has two texts that must NOT be the same:\n\nEXPRESSION — what a person would say to single out the object or objects. It may describe motion, temporal order, interaction with other objects, and spatial relations. It should read like something a person would actually say.\n  Good: \"the black car that's turning left\"\n  Good: \"the cyclist who overtakes the bus\"\n  Good: \"the two people who walk out of frame on the right\"\n\nNOUN_PHRASE — the same object as a plain, timeless visual category with its visible attributes, and NOTHING else. The detector is prompted with this text and looks at ONE frame at a time, so anything about time, order, or action is useless to it and makes it fail.\n  Allowed: category nouns, colours, materials, clothing, size, type\n  Forbidden: verbs and participles of action (\"turning\", \"walks\", \"is moving\"), temporal words (\"then\", \"starts\", \"after\", \"first\", \"last\"), relative clauses (\"that ...\", \"which ...\", \"who ...\"), ordinals and rank words (\"first\", \"leftmost\", \"nearest\"), and anything over six words.\n  \"the black car that's turning left\" -> \"black car\"\n  \"the cyclist who overtakes the bus\" -> \"cyclist\"\n  \"the two people who walk out of frame on the right\" -> \"person\"\n\nKIND — what makes the expression hard: \"motion\", \"appearance\", \"spatial\", or \"interaction\". Aim for about 35% \"motion\" items: motion is the part a single-frame model cannot already solve, so it is the most valuable, but a list of only motion items is not usable training data either.\n\nEXPECTED_COUNT — how many distinct objects in the clip match the EXPRESSION. Count carefully; this is checked against the detector's answer, and a wrong count discards the item.\n\nIS_NEGATIVE — at most ONE item may be a negative, and only when you also write at least one non-negative item. A negative is an expression that recombines attributes that ARE present in the clip into a whole that is NOT. If the only black car is moving, \"the parked black car\" is a good negative. Its NOUN_PHRASE must still be the category that IS present (\"black car\"), because that is what makes the absence checkable rather than merely unverified. A negative's EXPECTED_COUNT is 0.\n\nRules that discard an item, so follow them exactly:\n- Never write coordinates, box values, pixel positions, or percentages of the frame.\n- Never let a NOUN_PHRASE contain a verb of action, a temporal word, a relative clause, or an ordinal.\n- Never write an EXPRESSION whose object is not visible in the clip, unless it is marked IS_NEGATIVE.\n- Prefer objects that are clearly visible and identifiable over background clutter.\n- Do not describe the camera or the shot; describe objects in the scene.\n","user_prompt":"Watch this clip and write between 1 and 3 items following the rules. Cover different objects rather than restating one object several ways, and include any negative in the same list.\n"},"outputs":{"raw_response":"{\n  \"items\": [\n    {\n      \"expression\": \"the white information sign posted near the stone wall\",\n      \"noun_phrase\": \"white sign\",\n      \"kind\": \"spatial\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the large boulders piled up against the blue sky\",\n      \"noun_phrase\": \"large boulder\",\n      \"kind\": \"appearance\",\n      \"expected_count\": 5,\n      \"is_negative\": false\n    }\n  ]\n}","raw_items":[{"expression":"the white information sign posted near the stone wall","noun_phrase":"white sign","kind":"spatial","expected_count":1,"is_negative":false},{"expression":"the large boulders piled up against the blue sky","noun_phrase":"large boulder","kind":"appearance","expected_count":5,"is_negative":false}],"accepted":[{"expression":"the white information sign posted near the stone wall","noun_phrase":"white sign","kind":"spatial","expected_count":1,"is_negative":false},{"expression":"the large boulders piled up against the blue sky","noun_phrase":"large boulder","kind":"appearance","expected_count":5,"is_negative":false}],"gated":[]}},"track":{"inputs":{"tracker":"Sam31MultiplexTracker","model_id":"/mnt/ceph3/ec/jacky/sam3.1","noun_phrases":["large boulder","white sign"],"expressions_in":2,"frames":63,"sample_fps":4.0,"frame_size":"1920x1080","timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88,10.14,10.4,10.66,10.92,11.18,11.44,11.700000000000001,11.96,12.22,12.48,12.74,13.0,13.26,13.52,13.780000000000001,14.040000000000001,14.3,14.56,14.82,15.08,15.34,15.6,15.860000000000001,16.12],"session":"one session over 63 JPEG frames; then per phrase: reset_session, add_prompt(text) on frame 0, propagate_in_video","passes":2},"outputs":{"raw_tracks":[{"noun_phrase":"large boulder","object_id":0,"observations":[{"timestamp":0.0,"box":[0.17343750596046448,0.24722221493721008,0.35729168355464935,0.4824073910713196],"occluded":false},{"timestamp":0.26,"box":[0.17291668057441711,0.2944444417953491,0.36093752086162567,0.52407406270504],"occluded":false},{"timestamp":0.52,"box":[0.17604167759418488,0.260185182094574,0.3630208522081375,0.5018518418073654],"occluded":false},{"timestamp":0.78,"box":[0.04635417088866234,0.22870369255542755,0.3645833395421505,0.470370352268219],"occluded":false},{"timestamp":1.04,"box":[0.06041666865348816,0.23240740597248077,0.35572919249534607,0.47870369255542755],"occluded":false},{"timestamp":1.3,"box":[0.12031250447034836,0.19074073433876038,0.3255208507180214,0.4749999940395355],"occluded":false},{"timestamp":1.56,"box":[0.06041666865348816,0.14444443583488464,0.6302083432674408,0.8796295821666718],"occluded":false},{"timestamp":1.82,"box":[0.12708334624767303,0.2796296179294586,0.3307291865348816,0.5583333075046539],"occluded":false},{"timestamp":2.08,"box":[0.1302083432674408,0.2768518328666687,0.7270833551883698,0.9148147702217102],"occluded":false},{"timestamp":2.34,"box":[0.10208334028720856,0.2768518328666687,0.8104167133569717,0.6990740597248077],"occluded":false},{"timestamp":2.6,"box":[0.19895833730697632,0.27314814925193787,0.4140625149011612,0.5259259343147278],"occluded":false},{"timestamp":2.86,"box":[0.21302084624767303,0.012037036940455437,0.9911458939313889,0.5462962966412306],"occluded":false},{"timestamp":3.12,"box":[0.25468751788139343,0.3055555522441864,0.7552084028720856,0.5611110925674438],"occluded":false},{"timestamp":3.38,"box":[0.25468751788139343,0.10925925523042679,0.9911458790302277,0.5703703537583351],"occluded":false},{"timestamp":3.64,"box":[0.23125001788139343,0.345370352268219,0.9911458790302277,0.6787036657333374],"occluded":false},{"timestamp":3.9,"box":[0.004687500186264515,0.35185185074806213,0.39479167480021715,0.7018518447875977],"occluded":false},{"timestamp":4.16,"box":[0.2223958522081375,0.06759259104728699,0.9911458641290665,0.7111110985279083],"occluded":false},{"timestamp":4.42,"box":[0.22343751788139343,0.06759259104728699,0.9911458790302277,0.7018518149852753],"occluded":false},{"timestamp":4.68,"box":[0.06041666865348816,0.06759259104728699,0.9911458790302277,0.8379629552364349],"occluded":false},{"timestamp":4.94,"box":[0.004687500186264515,0.06759259104728699,0.9911458613350987,0.6888888776302338],"occluded":false},{"timestamp":5.2,"box":[0.0182291679084301,0.06759259104728699,0.9911458902060986,0.6749999821186066],"occluded":false},{"timestamp":5.46,"box":[0.03281250223517418,0.06759259104728699,0.9911458753049374,0.7018518149852753],"occluded":false},{"timestamp":5.72,"box":[0.0468750037252903,0.31018519401550293,0.40260419622063637,0.6814814805984497],"occluded":false},{"timestamp":5.98,"box":[0.03281250223517418,0.26851850748062134,0.3984375260770321,0.662962943315506],"occluded":false},{"timestamp":6.24,"box":[0.11614584177732468,0.28240740299224854,0.3682292029261589,0.6231481432914734],"occluded":false},{"timestamp":6.5,"box":[0.1692708432674408,0.3203703761100769,0.7552084028720856,0.6333333253860474],"occluded":false},{"timestamp":6.76,"box":[0.0885416716337204,0.33796295523643494,0.41718752682209015,0.6675925850868225],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.18958334624767303,0.39629629254341125,0.41718752682209015,0.6861110925674438],"occluded":false},{"timestamp":7.28,"box":[0.17812500894069672,0.45277777314186096,0.7828125208616257,0.767592579126358],"occluded":false},{"timestamp":7.54,"box":[0.01875000074505806,0.40740740299224854,0.8661458976566792,0.7685185074806213],"occluded":false},{"timestamp":7.8,"box":[0.20364584028720856,0.4731481373310089,0.4364583492279053,0.7555555403232574],"occluded":false},{"timestamp":8.06,"box":[0.15781250596046448,0.5231481194496155,0.6270833611488342,0.7833333015441895],"occluded":false},{"timestamp":8.32,"box":[0.1432291716337204,0.36574074625968933,0.9354166835546494,0.7462962865829468],"occluded":false},{"timestamp":8.58,"box":[0.14375001192092896,0.39351850748062134,0.9354166984558105,0.7435185015201569],"occluded":false},{"timestamp":8.84,"box":[0.12916667759418488,0.49537035822868347,0.49010418355464935,0.7611110806465149],"occluded":false},{"timestamp":9.1,"box":[0.01875000074505806,0.4277777671813965,0.45625003054738045,0.7333333194255829],"occluded":false},{"timestamp":9.36,"box":[0.03229166939854622,0.3722222149372101,0.42916669324040413,0.7416666448116302],"occluded":false},{"timestamp":9.620000000000001,"box":[0.0,0.3324074149131775,0.5989583730697632,0.7731481492519379],"occluded":false},{"timestamp":9.88,"box":[0.0,0.3018518388271332,0.6411458849906921,0.7972221970558167],"occluded":false},{"timestamp":10.14,"box":[0.0,0.23981481790542603,0.5744792222976685,0.7592592239379883],"occluded":false},{"timestamp":10.4,"box":[0.0,0.2796296179294586,0.99114590883255,0.7805555164813995],"occluded":false},{"timestamp":10.66,"box":[0.0,0.2648148238658905,0.99114590883255,0.7759259045124054],"occluded":false},{"timestamp":10.92,"box":[0.0,0.25,0.612500011920929,0.7611110806465149],"occluded":false},{"timestamp":11.18,"box":[0.0,0.22499999403953552,0.5848958492279053,0.737962931394577],"occluded":false},{"timestamp":11.44,"box":[0.0,0.22685185074806213,0.5854166746139526,0.7333333194255829],"occluded":false},{"timestamp":11.700000000000001,"box":[0.0,0.25648146867752075,0.5848958492279053,0.7638888955116272],"occluded":false},{"timestamp":11.96,"box":[0.0,0.26851850748062134,0.5994791984558105,0.7740740776062012],"occluded":false},{"timestamp":12.22,"box":[0.0,0.26851850748062134,0.5572916865348816,0.7842592597007751],"occluded":false},{"timestamp":12.48,"box":[0.0,0.3009259104728699,0.28697916865348816,0.7916666567325592],"occluded":false},{"timestamp":12.74,"box":[0.0,0.28240740299224854,0.5989583730697632,0.7972221970558167],"occluded":false},{"timestamp":13.0,"box":[0.0,0.28240740299224854,0.2958333492279053,0.7944444417953491],"occluded":false},{"timestamp":13.26,"box":[0.0,0.29629629850387573,0.28541669249534607,0.8101851940155029],"occluded":false},{"timestamp":13.52,"box":[0.0,0.33703702688217163,0.557812511920929,0.8425925970077515],"occluded":false},{"timestamp":13.780000000000001,"box":[0.0,0.3157407343387604,0.5057291984558105,0.8361110985279083],"occluded":false},{"timestamp":14.040000000000001,"box":[0.0,0.3055555522441864,0.47812503576278687,0.8305555284023285],"occluded":false},{"timestamp":14.3,"box":[0.0,0.31203702092170715,0.557812511920929,0.8398148119449615],"occluded":false},{"timestamp":14.56,"box":[0.0,0.3194444477558136,0.5744792222976685,0.837037056684494],"occluded":false},{"timestamp":14.82,"box":[0.0,0.32777777314186096,0.5848958492279053,0.893518477678299],"occluded":false},{"timestamp":15.08,"box":[0.0,0.32314813137054443,0.5989583730697632,0.9148147702217102],"occluded":false},{"timestamp":15.34,"box":[0.0,0.24444444477558136,0.6130208373069763,0.7962963134050369],"occluded":false},{"timestamp":15.6,"box":[0.004166666883975267,0.24074073135852814,0.6130208414979279,0.8092592507600784],"occluded":false},{"timestamp":15.860000000000001,"box":[0.0078125,0.25925925374031067,0.5989583730697632,0.6564814746379852],"occluded":false},{"timestamp":16.12,"box":[0.0442708358168602,0.2898148000240326,0.5437500402331352,0.6453703343868256],"occluded":false}],"scores":[0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035]},{"noun_phrase":"large boulder","object_id":1,"observations":[{"timestamp":0.0,"box":[0.6536458730697632,0.032407406717538834,0.9994792342185974,0.9666666425764561],"occluded":false},{"timestamp":0.26,"box":[0.6484375596046448,0.031481482088565826,0.9994792342185974,0.9333333149552345],"occluded":false},{"timestamp":0.52,"box":[0.6760417222976685,0.03055555559694767,0.9994792342185974,0.9574073646217585],"occluded":false},{"timestamp":0.78,"box":[0.06041666865348816,0.03055555559694767,0.999479204416275,0.9037036988884211],"occluded":false},{"timestamp":1.04,"box":[0.14375001192092896,0.03055555559694767,0.9994792342185974,0.8759259078651667],"occluded":false},{"timestamp":1.3,"box":[0.11614584177732468,0.031481482088565826,0.9994792267680168,0.9277777448296547],"occluded":false},{"timestamp":1.56,"box":[0.22708334028720856,0.03055555559694767,0.9994791895151138,0.9629629347473383],"occluded":false},{"timestamp":1.82,"box":[0.5880208611488342,0.03055555559694767,0.999479204416275,0.8925925586372614],"occluded":false},{"timestamp":2.08,"box":[0.26875001192092896,0.03055555559694767,0.9994792342185974,0.914814779534936],"occluded":false},{"timestamp":2.34,"box":[0.29635417461395264,0.03055555559694767,0.9994792342185974,0.5462963078171015],"occluded":false},{"timestamp":2.6,"box":[0.9458333849906921,0.031481482088565826,0.9994792193174362,0.18703702837228775],"occluded":false},{"timestamp":3.9,"box":[0.9895833730697632,0.031481482088565826,0.9994792072102427,0.06296296417713165],"occluded":false},{"timestamp":15.860000000000001,"box":[0.004687500186264515,0.28240740299224854,0.9994792463257909,0.970370352268219],"occluded":false},{"timestamp":16.12,"box":[0.07343750447034836,0.19907407462596893,0.9994792267680168,0.9722221940755844],"occluded":false}],"scores":[0.9179486632347107,0.9179486632347107,0.9179486632347107,0.9179486632347107,0.9179486632347107,0.9179486632347107,0.9179486632347107,0.9179486632347107,0.9179486632347107,0.9179486632347107,0.9179486632347107,0.9179486632347107,0.9179486632347107,0.9179486632347107]},{"noun_phrase":"large boulder","object_id":2,"observations":[{"timestamp":0.0,"box":[0.2343750149011612,0.21574074029922485,0.3192708492279053,0.33888888359069824],"occluded":false},{"timestamp":0.26,"box":[0.23489584028720856,0.2629629671573639,0.3203125149011612,0.3787037059664726],"occluded":false},{"timestamp":0.52,"box":[0.24062500894069672,0.2351851761341095,0.3265625163912773,0.35462962090969086],"occluded":false},{"timestamp":0.78,"box":[0.24270834028720856,0.20277777314186096,0.3291666805744171,0.32592591643333435],"occluded":false},{"timestamp":1.04,"box":[0.14375001192092896,0.21203702688217163,0.3239583522081375,0.36574073135852814],"occluded":false},{"timestamp":1.3,"box":[0.20104168355464935,0.1814814805984497,0.2916666865348816,0.3157407343387604],"occluded":false},{"timestamp":1.56,"box":[0.18020834028720856,0.14722222089767456,0.27395834773778915,0.2888888865709305],"occluded":false},{"timestamp":1.82,"box":[0.20937500894069672,0.27407407760620117,0.2968750149011612,0.4018518477678299],"occluded":false},{"timestamp":2.08,"box":[0.15781250596046448,0.26759257912635803,0.33802084624767303,0.42129628360271454],"occluded":false},{"timestamp":2.34,"box":[0.18541668355464935,0.27314814925193787,0.35833336412906647,0.42129629850387573],"occluded":false},{"timestamp":2.6,"box":[0.2942708432674408,0.2638888955116272,0.37552084773778915,0.38055556267499924],"occluded":false},{"timestamp":2.86,"box":[0.29322919249534607,0.060185182839632034,0.38333336263895035,0.3749999888241291],"occluded":false},{"timestamp":3.12,"box":[0.3489583432674408,0.3018518388271332,0.4270833507180214,0.41018517315387726],"occluded":false},{"timestamp":3.38,"box":[0.27916666865348816,0.31296294927597046,0.42500001192092896,0.42129628360271454],"occluded":false},{"timestamp":3.64,"box":[0.3114583492279053,0.41018518805503845,0.3906250223517418,0.5240740776062012],"occluded":false},{"timestamp":3.9,"box":[0.20937500894069672,0.4305555522441864,0.36250001192092896,0.5472222194075584],"occluded":false},{"timestamp":4.16,"box":[0.31041666865348816,0.4407407343387604,0.38854167610406876,0.5555555447936058],"occluded":false},{"timestamp":4.42,"box":[0.3140625059604645,0.4342592656612396,0.6854166984558105,0.7268518507480621],"occluded":false},{"timestamp":4.68,"box":[0.23750001192092896,0.43796294927597046,0.38750001788139343,0.5638888776302338],"occluded":false},{"timestamp":4.94,"box":[0.2802083492279053,0.4203703701496124,0.35989585518836975,0.5398148149251938],"occluded":false},{"timestamp":5.2,"box":[0.28697916865348816,0.40925925970077515,0.36770834028720856,0.5277777761220932],"occluded":false},{"timestamp":5.46,"box":[0.29322919249534607,0.4342592656612396,0.3744791969656944,0.555555559694767],"occluded":false},{"timestamp":5.72,"box":[0.29895836114883423,0.4203703701496124,0.3796875327825546,0.5407407358288765],"occluded":false},{"timestamp":5.98,"box":[0.30416667461395264,0.39814814925193787,0.3812500089406967,0.5222222208976746],"occluded":false},{"timestamp":6.24,"box":[0.27395835518836975,0.3472222089767456,0.35364586114883423,0.47499997913837433],"occluded":false},{"timestamp":6.5,"box":[0.28958335518836975,0.36944442987442017,0.3697916939854622,0.48981479555368423],"occluded":false},{"timestamp":6.76,"box":[0.3125000298023224,0.4148148000240326,0.39479169994592667,0.5305555388331413],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.20937500894069672,0.4175925850868225,0.3916666805744171,0.5472222119569778],"occluded":false},{"timestamp":7.28,"box":[0.30416667461395264,0.5027777552604675,0.385416679084301,0.6222222000360489],"occluded":false},{"timestamp":7.54,"box":[0.29114586114883423,0.4972222149372101,0.37395836412906647,0.6175925806164742],"occluded":false},{"timestamp":7.8,"box":[0.32760417461395264,0.5074074268341064,0.4114583432674408,0.6166666820645332],"occluded":false},{"timestamp":8.06,"box":[0.008333333767950535,0.5425925850868225,0.45052085164934397,0.645370364189148],"occluded":false},{"timestamp":8.32,"box":[0.38333335518836975,0.5185185074806213,0.46822918951511383,0.6138888746500015],"occluded":false},{"timestamp":8.58,"box":[0.3838541805744171,0.5138888955116272,0.4677083492279053,0.6120370402932167],"occluded":false},{"timestamp":8.84,"box":[0.008333333767950535,0.5046296119689941,0.4567708456888795,0.6231481283903122],"occluded":false},{"timestamp":9.1,"box":[0.18541668355464935,0.44907405972480774,0.42083336412906647,0.5777777582406998],"occluded":false},{"timestamp":9.36,"box":[0.03281250223517418,0.4120370149612427,0.38593750819563866,0.5546296089887619],"occluded":false},{"timestamp":9.620000000000001,"box":[0.20989584922790527,0.3861111104488373,0.35312502086162567,0.5555555522441864],"occluded":false},{"timestamp":9.88,"box":[0.1640625149011612,0.36759257316589355,0.325520858168602,0.5601851642131805],"occluded":false},{"timestamp":10.14,"box":[0.11354167014360428,0.29999998211860657,0.6020833626389503,0.5120370090007782],"occluded":false},{"timestamp":10.4,"box":[0.06822916865348816,0.2925925850868225,0.25468751788139343,0.6370370090007782],"occluded":false},{"timestamp":10.66,"box":[0.008333333767950535,0.2648148238658905,0.25520835164934397,0.6157407462596893],"occluded":false},{"timestamp":10.92,"box":[0.05260416865348816,0.24537037312984467,0.2552083432674408,0.581481471657753],"occluded":false},{"timestamp":11.18,"box":[0.0,0.2231481373310089,0.25729167461395264,0.6046296060085297],"occluded":false},{"timestamp":11.44,"box":[0.04322917014360428,0.21851851046085358,0.504687525331974,0.460185170173645],"occluded":false},{"timestamp":11.700000000000001,"box":[0.05781250447034836,0.2537037134170532,0.2562500163912773,0.49074074625968933],"occluded":false},{"timestamp":11.96,"box":[0.04270833730697632,0.25925925374031067,0.2421875149011612,0.4999999850988388],"occluded":false},{"timestamp":12.22,"box":[0.03489583358168602,0.2666666507720947,0.23541667684912682,0.5092592388391495],"occluded":false},{"timestamp":12.48,"box":[0.03437500074505806,0.27407407760620117,0.23489584401249886,0.5175925940275192],"occluded":false},{"timestamp":12.74,"box":[0.04375000298023224,0.2787036895751953,0.24322918057441711,0.5203703492879868],"occluded":false},{"timestamp":13.0,"box":[0.0468750037252903,0.2759259343147278,0.9770833887159824,0.7407407462596893],"occluded":false},{"timestamp":13.26,"box":[0.0364583358168602,0.19907407462596893,0.45625003427267075,0.5324073880910873],"occluded":false},{"timestamp":13.52,"box":[0.03750000149011612,0.22685185074806213,0.48906252533197403,0.5629629492759705],"occluded":false},{"timestamp":13.780000000000001,"box":[0.04791666939854622,0.22685185074806213,0.40833334997296333,0.5583333373069763],"occluded":false},{"timestamp":14.040000000000001,"box":[0.04895833507180214,0.22685185074806213,0.42916667833924294,0.5509259104728699],"occluded":false},{"timestamp":14.3,"box":[0.05052083730697632,0.22685185074806213,0.421354204416275,0.5601851642131805],"occluded":false},{"timestamp":14.56,"box":[0.05312500149011612,0.24074073135852814,0.39375001937150955,0.5620370358228683],"occluded":false},{"timestamp":14.82,"box":[0.06093750149011612,0.24074073135852814,0.3802083507180214,0.5574073940515518],"occluded":false},{"timestamp":15.08,"box":[0.07760416716337204,0.23981481790542603,0.56041669100523,0.5555555522441864],"occluded":false},{"timestamp":15.34,"box":[0.1328125,0.22685185074806213,0.5744791924953461,0.5194444358348846],"occluded":false},{"timestamp":15.6,"box":[0.1744791716337204,0.3083333373069763,0.49895836412906647,0.4861111044883728],"occluded":false},{"timestamp":15.860000000000001,"box":[0.2031250149011612,0.22685185074806213,0.4697916954755783,0.4703703671693802],"occluded":false},{"timestamp":16.12,"box":[0.23125001788139343,0.25462961196899414,0.5604166984558105,0.47499997913837433],"occluded":false}],"scores":[0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156,0.7655677795410156]},{"noun_phrase":"large boulder","object_id":3,"observations":[{"timestamp":0.0,"box":[0.04062500223517418,0.1537037044763565,0.20781251415610313,0.375],"occluded":false},{"timestamp":0.26,"box":[0.04218750074505806,0.21666666865348816,0.20364584401249886,0.4287036955356598],"occluded":false},{"timestamp":0.52,"box":[0.04843750223517418,0.17037037014961243,0.20625000819563866,0.3879629522562027],"occluded":false},{"timestamp":0.78,"box":[0.04635417088866234,0.138888880610466,0.2067708484828472,0.36574073135852814],"occluded":false},{"timestamp":1.04,"box":[0.03750000149011612,0.14166666567325592,0.19843750447034836,0.36759258806705475],"occluded":false},{"timestamp":1.3,"box":[0.0,0.07037036865949631,0.15052084624767303,0.32685183733701706],"occluded":false},{"timestamp":1.56,"box":[0.0,0.032407406717538834,0.12864583730697632,0.2898148037493229],"occluded":false},{"timestamp":1.82,"box":[0.0,0.1814814805984497,0.14739583432674408,0.42592592537403107],"occluded":false},{"timestamp":2.08,"box":[0.05937500298023224,0.20000000298023224,0.19739584624767303,0.4148148149251938],"occluded":false},{"timestamp":2.34,"box":[0.09062500298023224,0.19907407462596893,0.21979168057441711,0.42129629850387573],"occluded":false},{"timestamp":2.6,"box":[0.11145833879709244,0.21759258210659027,0.23541668057441711,0.4166666567325592],"occluded":false},{"timestamp":2.86,"box":[0.12135417014360428,0.22407406568527222,0.24322917312383652,0.4287036955356598],"occluded":false},{"timestamp":3.12,"box":[0.18333333730697632,0.25740739703178406,0.7828125357627869,0.5324074029922485],"occluded":false},{"timestamp":3.38,"box":[0.18020834028720856,0.27222222089767456,0.7692708522081375,0.5185185074806213],"occluded":false},{"timestamp":3.64,"box":[0.1354166716337204,0.35370370745658875,0.24895834177732468,0.5444444417953491],"occluded":false},{"timestamp":3.9,"box":[0.0911458358168602,0.36759257316589355,0.21458334475755692,0.5685184895992279],"occluded":false},{"timestamp":4.16,"box":[0.13125000894069672,0.3759259283542633,0.755208358168602,0.6712962985038757],"occluded":false},{"timestamp":4.42,"box":[0.13229167461395264,0.06759259104728699,0.99114590883255,0.6574073731899261],"occluded":false},{"timestamp":4.68,"box":[0.12552084028720856,0.06759259104728699,0.9911458641290665,0.6851851642131805],"occluded":false},{"timestamp":4.94,"box":[0.08281250298023224,0.3333333134651184,0.20468750596046448,0.5370370149612427],"occluded":false},{"timestamp":5.2,"box":[0.0911458358168602,0.32222220301628113,0.755208395421505,0.6574073731899261],"occluded":false},{"timestamp":5.46,"box":[0.07447917014360428,0.06759259104728699,0.9911458566784859,0.6574073731899261],"occluded":false},{"timestamp":5.72,"box":[0.10572917014360428,0.3314814865589142,0.7557292059063911,0.6435185074806213],"occluded":false},{"timestamp":5.98,"box":[0.11093750596046448,0.28703704476356506,0.22291667759418488,0.49166667461395264],"occluded":false},{"timestamp":6.24,"box":[0.0651041716337204,0.20370370149612427,0.18802084773778915,0.4296296238899231],"occluded":false},{"timestamp":6.5,"box":[0.07916667312383652,0.2666666507720947,0.5609375461935997,0.5185185074806213],"occluded":false},{"timestamp":6.76,"box":[0.11302083730697632,0.33796295523643494,0.22239584475755692,0.527777761220932],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.10468750447034836,0.3712962865829468,0.21510417759418488,0.5583333224058151],"occluded":false},{"timestamp":7.28,"box":[0.10312500596046448,0.42129629850387573,0.20937500894069672,0.6157407462596893],"occluded":false},{"timestamp":7.54,"box":[0.08489583432674408,0.402777761220932,0.19479167461395264,0.6009259074926376],"occluded":false},{"timestamp":7.8,"box":[0.13072916865348816,0.460185170173645,0.22968750447034836,0.6333333104848862],"occluded":false},{"timestamp":8.06,"box":[0.18020834028720856,0.5268518328666687,0.268229179084301,0.6935184895992279],"occluded":false},{"timestamp":8.32,"box":[0.1953125149011612,0.5249999761581421,0.28593751788139343,0.6620370149612427],"occluded":false},{"timestamp":8.58,"box":[0.19479167461395264,0.5074074268341064,0.2880208492279053,0.6620370596647263],"occluded":false},{"timestamp":8.84,"box":[0.18072918057441711,0.5,0.27135418355464935,0.6648148149251938],"occluded":false},{"timestamp":9.1,"box":[0.08645834028720856,0.432407408952713,0.20104167610406876,0.6240740716457367],"occluded":false},{"timestamp":9.36,"box":[0.0,0.3759259283542633,0.11927083879709244,0.6203703731298447],"occluded":false},{"timestamp":9.620000000000001,"box":[0.0,0.35185185074806213,0.05000000447034836,0.6361111104488373],"occluded":false},{"timestamp":15.860000000000001,"box":[0.0,0.21203702688217163,0.9963542222976685,0.32592591643333435],"occluded":false},{"timestamp":16.12,"box":[0.0,0.2916666567325592,0.0833333358168602,0.534259244799614],"occluded":false}],"scores":[0.6952381134033203,0.6952381134033203,0.6952381134033203,0.6952381134033203,0.6952381134033203,0.6952381134033203,0.6952381134033203,0.6952381134033203,0.6952381134033203,0.6952381134033203,0.6952381134033203,0.6952381134033203,0.6952381134033203,0.6952381134033203,0.6952381134033203,0.6952381134033203,0.6952381134033203,0.6952381134033203,0.6952381134033203,0.6952381134033203,0.6952381134033203,0.6952381134033203,0.6952381134033203,0.6952381134033203,0.6952381134033203,0.6952381134033203,0.6952381134033203,0.6952381134033203,0.6952381134033203,0.6952381134033203,0.6952381134033203,0.6952381134033203,0.6952381134033203,0.6952381134033203,0.6952381134033203,0.6952381134033203,0.6952381134033203,0.6952381134033203,0.6952381134033203,0.6952381134033203]},{"noun_phrase":"large boulder","object_id":4,"observations":[{"timestamp":0.0,"box":[0.33645835518836975,0.18518517911434174,0.4802083671092987,0.31018517911434174],"occluded":false},{"timestamp":0.26,"box":[0.33385419845581055,0.2231481373310089,0.47968754172325134,0.34166665375232697],"occluded":false},{"timestamp":0.52,"box":[0.3411458432674408,0.20555555820465088,0.4869791865348816,0.3305555582046509],"occluded":false},{"timestamp":0.78,"box":[0.34427085518836975,0.17222221195697784,0.49166668951511383,0.3009259104728699],"occluded":false},{"timestamp":1.04,"box":[0.3395833373069763,0.1814814805984497,0.48750001192092896,0.3157407343387604],"occluded":false},{"timestamp":1.3,"box":[0.3083333373069763,0.17037037014961243,0.4598958492279053,0.31481480598449707],"occluded":false},{"timestamp":1.56,"box":[0.29114586114883423,0.14629629254341125,0.4453125298023224,0.3018518388271332],"occluded":false},{"timestamp":1.82,"box":[0.3151041865348816,0.25462961196899414,0.536458358168602,0.5601851642131805],"occluded":false},{"timestamp":2.08,"box":[0.3453125059604645,0.23888888955116272,0.502083346247673,0.3740740716457367],"occluded":false},{"timestamp":2.34,"box":[0.37552085518836975,0.2490740716457367,0.520833358168602,0.38055555522441864],"occluded":false},{"timestamp":2.6,"box":[0.3916666805744171,0.2222222238779068,0.5390625149011612,0.3462962955236435],"occluded":false},{"timestamp":2.86,"box":[0.31041666865348816,0.21944443881511688,0.5468750149011612,0.33888888359069824],"occluded":false},{"timestamp":3.12,"box":[0.008333333767950535,0.17129629850387573,0.7552083814516664,0.5185185074806213],"occluded":false},{"timestamp":3.38,"box":[0.4427083432674408,0.27037036418914795,0.5901041775941849,0.39537036418914795],"occluded":false},{"timestamp":3.64,"box":[0.4072916805744171,0.3759259283542633,0.6890625357627869,0.7518518567085266],"occluded":false},{"timestamp":3.9,"box":[0.38072919845581055,0.39907407760620117,0.5744792073965073,0.7824074029922485],"occluded":false},{"timestamp":4.16,"box":[0.40781253576278687,0.06759259104728699,0.99114590883255,0.5388888716697693],"occluded":false},{"timestamp":4.42,"box":[0.41197919845581055,0.4009259045124054,0.7000000476837158,0.7268518209457397],"occluded":false},{"timestamp":4.68,"box":[0.06041666865348816,0.3314814865589142,0.9911458790302277,0.5509259253740311],"occluded":false},{"timestamp":4.94,"box":[0.38229167461395264,0.4000000059604645,0.5312500149011612,0.5333333313465118],"occluded":false},{"timestamp":5.2,"box":[0.39010417461395264,0.3888888955116272,0.5401041805744171,0.5250000059604645],"occluded":false},{"timestamp":5.46,"box":[0.39635419845581055,0.4120370149612427,0.5604167133569717,0.9490740299224854],"occluded":false},{"timestamp":5.72,"box":[0.40052086114883423,0.4009259045124054,0.5604166984558105,0.8101851642131805],"occluded":false},{"timestamp":5.98,"box":[0.4052083492279053,0.39259257912635803,0.5541666895151138,0.5388888716697693],"occluded":false},{"timestamp":6.24,"box":[0.37812501192092896,0.35185185074806213,0.528645858168602,0.5074073970317841],"occluded":false},{"timestamp":6.5,"box":[0.3958333432674408,0.35462963581085205,0.5458333492279053,0.49537037312984467],"occluded":false},{"timestamp":6.76,"box":[0.41979169845581055,0.3861111104488373,0.5729167014360428,0.5175925940275192],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.4166666865348816,0.39444443583488464,0.5708333551883698,0.5212962925434113],"occluded":false},{"timestamp":7.28,"box":[0.41093751788139343,0.47685185074806213,0.5656250268220901,0.614814817905426],"occluded":false},{"timestamp":7.54,"box":[0.40000003576278687,0.47592592239379883,0.5562500506639481,0.6157407313585281],"occluded":false},{"timestamp":7.8,"box":[0.43697917461395264,0.4564814865589142,0.5927083492279053,0.5814814865589142],"occluded":false},{"timestamp":8.06,"box":[0.37968751788139343,0.4722222089767456,0.632291704416275,0.585185170173645],"occluded":false},{"timestamp":8.32,"box":[0.008333333767950535,0.4314814805984497,0.6510416949167848,0.5601851791143417],"occluded":false},{"timestamp":8.58,"box":[0.26875001192092896,0.39351850748062134,0.9354166984558105,0.5481481403112411],"occluded":false},{"timestamp":8.84,"box":[0.12968750298023224,0.44999998807907104,0.6453125029802322,0.5666666552424431],"occluded":false},{"timestamp":9.1,"box":[0.4489583671092987,0.38148146867752075,0.632291704416275,0.5194444358348846],"occluded":false},{"timestamp":9.36,"box":[0.03281250223517418,0.3203703761100769,0.6375000141561031,0.47870370745658875],"occluded":false},{"timestamp":9.620000000000001,"box":[0.3906250298023224,0.28703704476356506,0.9494792520999908,0.8518518507480621],"occluded":false},{"timestamp":9.88,"box":[0.3687500059604645,0.2620370388031006,0.6500000059604645,0.47870370745658875],"occluded":false},{"timestamp":10.14,"box":[0.04635417088866234,0.20000000298023224,0.9770833812654018,0.7129629403352737],"occluded":false},{"timestamp":10.4,"box":[0.14375001192092896,0.1944444477558136,0.6171875298023224,0.45277777314186096],"occluded":false},{"timestamp":10.66,"box":[0.04635417088866234,0.16388888657093048,0.6364583931863308,0.43518517911434174],"occluded":false},{"timestamp":10.92,"box":[0.3057291805744171,0.14074073731899261,0.6375000178813934,0.41018517315387726],"occluded":false},{"timestamp":11.18,"box":[0.14375001192092896,0.11759258806705475,0.6427083611488342,0.388888880610466],"occluded":false},{"timestamp":11.44,"box":[0.2984375059604645,0.11296296119689941,0.6302083432674408,0.38055554032325745],"occluded":false},{"timestamp":11.700000000000001,"box":[0.31041666865348816,0.14629629254341125,0.6411458551883698,0.412962943315506],"occluded":false},{"timestamp":11.96,"box":[0.31041666865348816,0.3046296238899231,0.628125011920929,0.4277777671813965],"occluded":false},{"timestamp":12.22,"box":[0.4932291805744171,0.31388887763023376,0.6208333522081375,0.43888887763023376],"occluded":false},{"timestamp":12.48,"box":[0.49270835518836975,0.32314813137054443,0.6203125268220901,0.44907405972480774],"occluded":false},{"timestamp":12.74,"box":[0.5,0.32777777314186096,0.628125011920929,0.45185184478759766],"occluded":false},{"timestamp":13.0,"box":[0.5020833611488342,0.32777777314186096,0.6291667073965073,0.45092591643333435],"occluded":false},{"timestamp":13.26,"box":[0.49375003576278687,0.34351852536201477,0.6203125417232513,0.46851852536201477],"occluded":false},{"timestamp":13.52,"box":[0.49375003576278687,0.3759259283542633,0.6208333820104599,0.5018518567085266],"occluded":false},{"timestamp":13.780000000000001,"box":[0.5010417103767395,0.3768518567085266,0.6291667222976685,0.5037037134170532],"occluded":false},{"timestamp":14.040000000000001,"box":[0.503125011920929,0.3722222149372101,0.6307291835546494,0.4962962865829468],"occluded":false},{"timestamp":14.3,"box":[0.31041666865348816,0.37870368361473083,0.6317708492279053,0.5046296119689941],"occluded":false},{"timestamp":14.56,"box":[0.5052083730697632,0.37962961196899414,0.6317708790302277,0.5027777552604675],"occluded":false},{"timestamp":14.82,"box":[0.31041666865348816,0.364814817905426,0.6192708611488342,0.4824074059724808],"occluded":false},{"timestamp":15.08,"box":[0.39375001192092896,0.24074073135852814,0.6125000268220901,0.4722222089767456],"occluded":false},{"timestamp":15.34,"box":[0.5135416984558105,0.3444444239139557,0.6234375387430191,0.44999998062849045],"occluded":false},{"timestamp":15.6,"box":[0.5213541984558105,0.317592591047287,0.6234375387430191,0.4138888865709305],"occluded":false},{"timestamp":15.860000000000001,"box":[0.5151041746139526,0.3046296238899231,0.6078125163912773,0.39537036418914795],"occluded":false},{"timestamp":16.12,"box":[0.5093750357627869,0.3305555582046509,0.5927083715796471,0.40833333134651184],"occluded":false}],"scores":[0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933]},{"noun_phrase":"large boulder","object_id":5,"observations":[{"timestamp":0.26,"box":[0.00364583358168602,0.3166666626930237,0.16302084550261497,0.45370370149612427],"occluded":false},{"timestamp":0.52,"box":[0.00729166716337204,0.2648148238658905,0.16302084177732468,0.4148148149251938],"occluded":false},{"timestamp":0.78,"box":[0.00729166716337204,0.2351851761341095,0.16718750447034836,0.3861110955476761],"occluded":false},{"timestamp":1.04,"box":[0.0,0.23981481790542603,0.15937501192092896,0.3898148089647293],"occluded":false},{"timestamp":1.3,"box":[0.0,0.18703703582286835,0.10729166865348816,0.35185185074806213],"occluded":false},{"timestamp":1.56,"box":[0.0,0.14999999105930328,0.12656250596046448,0.591666653752327],"occluded":false},{"timestamp":1.82,"box":[0.0,0.28425925970077515,0.11510417610406876,0.446296289563179],"occluded":false},{"timestamp":2.08,"box":[0.006770833861082792,0.2888888716697693,0.1692708428017795,0.4425925761461258],"occluded":false},{"timestamp":2.34,"box":[0.03593750298023224,0.28148147463798523,0.19062501192092896,0.43611110746860504],"occluded":false},{"timestamp":2.6,"box":[0.06302084028720856,0.26851850748062134,0.21666668355464935,0.4462962746620178],"occluded":false},{"timestamp":2.86,"box":[0.04531250149011612,0.21944443881511688,0.22447917610406876,0.44999998807907104],"occluded":false},{"timestamp":3.12,"box":[0.13906250894069672,0.32777777314186096,0.27031251788139343,0.4740740656852722],"occluded":false},{"timestamp":3.38,"box":[0.13333334028720856,0.3499999940395355,0.26875001192092896,0.4879629611968994],"occluded":false},{"timestamp":3.64,"box":[0.0598958358168602,0.4314814805984497,0.2317708507180214,0.7611111104488373],"occluded":false},{"timestamp":3.9,"box":[0.0,0.4037036895751953,0.2135416716337204,0.8101851642131805],"occluded":false},{"timestamp":4.16,"box":[0.0,0.42129629850387573,0.21041667461395264,0.7620370388031006],"occluded":false},{"timestamp":4.42,"box":[0.0,0.44907405972480774,0.2135416716337204,0.7824073731899261],"occluded":false},{"timestamp":4.68,"box":[0.05572916939854622,0.45185184478759766,0.20208333805203438,0.6055555492639542],"occluded":false},{"timestamp":4.94,"box":[0.0,0.39907407760620117,0.1614583432674408,0.5787037014961243],"occluded":false},{"timestamp":5.2,"box":[0.0,0.36574074625968933,0.17083334922790527,0.567592591047287],"occluded":false},{"timestamp":5.46,"box":[0.01770833507180214,0.4148148000240326,0.17500001564621925,0.595370352268219],"occluded":false},{"timestamp":5.72,"box":[0.02395833469927311,0.4000000059604645,0.17968750931322575,0.5694444477558136],"occluded":false},{"timestamp":5.98,"box":[0.004687500186264515,0.3166666626930237,0.17343750316649675,0.5249999910593033],"occluded":false},{"timestamp":6.24,"box":[0.0,0.24074073135852814,0.5604166984558105,0.7407407313585281],"occluded":false},{"timestamp":6.5,"box":[0.0,0.28333333134651184,0.1588541716337204,0.7129629552364349],"occluded":false},{"timestamp":6.76,"box":[0.03437500074505806,0.3759259283542633,0.18854166939854622,0.5814814865589142],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.02500000223517418,0.412962943315506,0.18385417386889458,0.6101851612329483],"occluded":false},{"timestamp":7.28,"box":[0.0,0.45092591643333435,0.16718751192092896,0.6611111015081406],"occluded":false},{"timestamp":7.54,"box":[0.0,0.4342592656612396,0.15156251192092896,0.6518518477678299],"occluded":false},{"timestamp":7.8,"box":[0.02239583432674408,0.5046296119689941,0.20156250894069672,0.6972222030162811],"occluded":false},{"timestamp":8.06,"box":[0.05000000447034836,0.5824074149131775,0.25208335369825363,0.7555555552244186],"occluded":false},{"timestamp":8.32,"box":[0.10572917014360428,0.5907407402992249,0.6854167059063911,0.7472222149372101],"occluded":false},{"timestamp":8.58,"box":[0.09843750298023224,0.5703703761100769,0.2718750089406967,0.7240740805864334],"occluded":false},{"timestamp":8.84,"box":[0.0703125,0.5666666626930237,0.25312501192092896,0.7277777642011642],"occluded":false},{"timestamp":9.1,"box":[0.0,0.4990740716457367,0.1796875149011612,0.7037037014961243],"occluded":false},{"timestamp":9.36,"box":[0.0,0.520370364189148,0.09010417014360428,0.7083333283662796],"occluded":false}],"scores":[0.675126850605011,0.675126850605011,0.675126850605011,0.675126850605011,0.675126850605011,0.675126850605011,0.675126850605011,0.675126850605011,0.675126850605011,0.675126850605011,0.675126850605011,0.675126850605011,0.675126850605011,0.675126850605011,0.675126850605011,0.675126850605011,0.675126850605011,0.675126850605011,0.675126850605011,0.675126850605011,0.675126850605011,0.675126850605011,0.675126850605011,0.675126850605011,0.675126850605011,0.675126850605011,0.675126850605011,0.675126850605011,0.675126850605011,0.675126850605011,0.675126850605011,0.675126850605011,0.675126850605011,0.675126850605011,0.675126850605011,0.675126850605011]},{"noun_phrase":"large boulder","object_id":6,"observations":[{"timestamp":1.3,"box":[0.5067708492279053,0.489814817905426,0.7229166924953461,0.9083333313465118],"occluded":false},{"timestamp":1.56,"box":[0.43385419249534607,0.47870370745658875,0.7072916924953461,0.9074074029922485],"occluded":false},{"timestamp":1.82,"box":[0.46666669845581055,0.5462962985038757,0.7645833790302277,0.970370352268219],"occluded":false},{"timestamp":2.08,"box":[0.3453125059604645,0.3583333194255829,0.8406250178813934,0.9694444239139557],"occluded":false},{"timestamp":2.34,"box":[0.5583333373069763,0.5361111164093018,0.8864583671092987,0.9712962806224823],"occluded":false},{"timestamp":2.6,"box":[0.19947917759418488,0.42129629850387573,0.9562500268220901,0.9666666388511658],"occluded":false},{"timestamp":2.86,"box":[0.6109375357627869,0.10833333432674408,0.9994792342185974,0.9990740865468979],"occluded":false},{"timestamp":3.12,"box":[0.6557291746139526,0.4833333194255829,0.999479204416275,0.9935185015201569],"occluded":false},{"timestamp":3.38,"box":[0.6473958492279053,0.12314814329147339,0.999479204416275,0.9925925731658936],"occluded":false},{"timestamp":3.64,"box":[0.3072916865348816,0.43518516421318054,0.9994792342185974,0.9990740716457367],"occluded":false},{"timestamp":3.9,"box":[0.24114584922790527,0.43518516421318054,0.9994792342185974,0.9990740716457367],"occluded":false},{"timestamp":4.16,"box":[0.3072916865348816,0.06666666269302368,0.9994792342185974,0.9990740418434143],"occluded":false},{"timestamp":4.42,"box":[0.6781250238418579,0.06666666269302368,0.999479204416275,0.9990740418434143],"occluded":false},{"timestamp":4.68,"box":[0.5151041746139526,0.3314814865589142,0.999479204416275,0.9990740716457367],"occluded":false},{"timestamp":4.94,"box":[0.6447917222976685,0.317592591047287,0.9994792342185974,0.9990740716457367],"occluded":false},{"timestamp":5.2,"box":[0.5015625357627869,0.317592591047287,0.9994792342185974,0.9990740716457367],"occluded":false},{"timestamp":5.46,"box":[0.29322919249534607,0.317592591047287,0.999479204416275,0.9907407462596893],"occluded":false},{"timestamp":5.72,"box":[0.6562500596046448,0.6370370388031006,0.9994792342185974,0.9759259223937988],"occluded":false},{"timestamp":5.98,"box":[0.6520833969116211,0.6481481194496155,0.9994792342185974,0.9990740418434143],"occluded":false},{"timestamp":6.24,"box":[0.5989583730697632,0.6212962865829468,0.9994792342185974,0.9740740656852722],"occluded":false},{"timestamp":6.5,"box":[0.6244791746139526,0.5694444179534912,0.999479204416275,0.9990740418434143],"occluded":false},{"timestamp":6.76,"box":[0.3072916865348816,0.4416666626930237,0.9994792342185974,0.9731481075286865],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.5291666984558105,0.5333333015441895,0.9994792342185974,0.9740740358829498],"occluded":false},{"timestamp":7.28,"box":[0.7500000596046448,0.6379629373550415,0.9994792342185974,0.9990740418434143],"occluded":false},{"timestamp":7.54,"box":[0.30677086114883423,0.5666666626930237,0.9994792342185974,0.9990740716457367],"occluded":false},{"timestamp":7.8,"box":[0.32083335518836975,0.5231481194496155,0.999479204416275,0.9722221791744232],"occluded":false},{"timestamp":8.06,"box":[0.45625001192092896,0.4546296298503876,0.9994792342185974,0.9990740716457367],"occluded":false},{"timestamp":8.32,"box":[0.612500011920929,0.33888888359069824,0.999479204416275,0.9990740418434143],"occluded":false},{"timestamp":8.58,"box":[0.9026042222976685,0.3592592477798462,0.9994792267680168,0.9731481075286865],"occluded":false},{"timestamp":8.84,"box":[0.22708334028720856,0.39629629254341125,0.9994791895151138,0.9759258925914764],"occluded":false},{"timestamp":9.1,"box":[0.18541668355464935,0.3481481373310089,0.9994792193174362,0.9712962806224823],"occluded":false},{"timestamp":9.36,"box":[0.98333340883255,0.4611110985279083,0.9994792435318232,0.7620370090007782],"occluded":false}],"scores":[0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563]},{"noun_phrase":"large boulder","object_id":7,"observations":[{"timestamp":1.82,"box":[0.28958335518836975,0.35462963581085205,0.5057291984558105,0.6314814686775208],"occluded":false},{"timestamp":2.08,"box":[0.14375001192092896,0.33888888359069824,0.6994792222976685,0.9148147702217102],"occluded":false},{"timestamp":2.34,"box":[0.17135417461395264,0.3481481373310089,0.5057291984558105,0.6249999701976776],"occluded":false},{"timestamp":2.6,"box":[0.19895833730697632,0.32592591643333435,0.5296875238418579,0.5916666388511658],"occluded":false},{"timestamp":2.86,"box":[0.21302084624767303,0.10925925523042679,0.9911458939313889,0.5879629626870155],"occluded":false},{"timestamp":3.12,"box":[0.24114584922790527,0.3629629611968994,0.7546875476837158,0.6370370388031006],"occluded":false},{"timestamp":3.38,"box":[0.25468751788139343,0.1787036955356598,0.9911458790302277,0.6444444358348846],"occluded":false},{"timestamp":3.64,"box":[0.32447919249534607,0.4749999940395355,0.5458333641290665,0.7583333253860474],"occluded":false},{"timestamp":3.9,"box":[0.14375001192092896,0.4962962865829468,0.5744791924953461,0.7962962687015533],"occluded":false},{"timestamp":4.16,"box":[0.004687500186264515,0.3314814865589142,0.9911458613350987,0.8101851940155029],"occluded":false},{"timestamp":4.42,"box":[0.38593751192092896,0.4981481432914734,0.6994791924953461,0.7962962985038757],"occluded":false},{"timestamp":4.68,"box":[0.07395833730697632,0.3314814865589142,0.99114590883255,0.8379629552364349],"occluded":false},{"timestamp":4.94,"box":[0.06041666865348816,0.317592591047287,0.9911458790302277,0.8101851642131805],"occluded":false},{"timestamp":5.2,"box":[0.01875000074505806,0.40740740299224854,0.8661458976566792,0.9212962985038757],"occluded":false},{"timestamp":5.46,"box":[0.03281250223517418,0.43518516421318054,0.589062537997961,0.9490740597248077],"occluded":false},{"timestamp":5.72,"box":[0.16770833730697632,0.49351850152015686,0.6437500417232513,0.8101851642131805],"occluded":false},{"timestamp":5.98,"box":[0.03229166939854622,0.37962961196899414,0.6020833440124989,0.8101851642131805],"occluded":false},{"timestamp":6.24,"box":[0.03281250223517418,0.35185185074806213,0.5473958514630795,0.7685185074806213],"occluded":false},{"timestamp":6.5,"box":[0.08802083879709244,0.3731481432914734,0.8942708745598793,0.8657407164573669],"occluded":false},{"timestamp":6.76,"box":[0.08802083879709244,0.43518516421318054,0.617187537252903,0.8796296119689941],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.38333335518836975,0.49166664481163025,0.7552083730697632,0.8101851642131805],"occluded":false},{"timestamp":7.28,"box":[0.15729168057441711,0.5712962746620178,0.616666704416275,0.9212962687015533],"occluded":false},{"timestamp":7.54,"box":[0.12916667759418488,0.5462962985038757,0.7687500268220901,0.8935185074806213],"occluded":false},{"timestamp":7.8,"box":[0.4046875238418579,0.5518518686294556,0.6447917073965073,0.9351851940155029],"occluded":false},{"timestamp":8.06,"box":[0.15781250596046448,0.5648148059844971,0.6994791924953461,0.8574073910713196],"occluded":false},{"timestamp":8.32,"box":[0.18541668355464935,0.36574074625968933,0.9354167431592941,0.8027777671813965],"occluded":false},{"timestamp":8.58,"box":[0.1848958432674408,0.39351850748062134,0.9354167282581329,0.8166666626930237],"occluded":false},{"timestamp":8.84,"box":[0.17135417461395264,0.5472221970558167,0.7276042103767395,0.8472221791744232],"occluded":false},{"timestamp":9.1,"box":[0.15781250596046448,0.49259257316589355,0.7411458790302277,0.8231481313705444],"occluded":false},{"timestamp":9.36,"box":[0.37968751788139343,0.45185184478759766,0.991666704416275,0.9351851642131805],"occluded":false},{"timestamp":9.620000000000001,"box":[0.10572917014360428,0.10185185074806213,0.996354229748249,0.9074073731899261],"occluded":false},{"timestamp":9.88,"box":[0.14375001192092896,0.09537036716938019,0.99114590883255,0.9500000029802322],"occluded":false},{"timestamp":10.14,"box":[0.11250000447034836,0.09537036716938019,0.9973958656191826,0.9527777582406998],"occluded":false},{"timestamp":10.4,"box":[0.03593750298023224,0.317592591047287,0.9973958879709244,0.9537036716938019],"occluded":false},{"timestamp":10.66,"box":[0.02916666865348816,0.33796295523643494,0.9963541924953461,0.9638888537883759],"occluded":false},{"timestamp":10.92,"box":[0.08802083879709244,0.345370352268219,0.9958333745598793,0.9462962746620178],"occluded":false},{"timestamp":11.18,"box":[0.01458333432674408,0.32314813137054443,0.6343750208616257,0.9194444417953491],"occluded":false},{"timestamp":11.44,"box":[0.0,0.3157407343387604,0.9494792222976685,0.9111110866069794],"occluded":false},{"timestamp":11.700000000000001,"box":[0.0,0.3490740656852722,0.9770833849906921,0.9472222328186035],"occluded":false},{"timestamp":11.96,"box":[0.03281250223517418,0.3611111044883728,0.9635417126119137,0.9611110687255859],"occluded":false},{"timestamp":12.22,"box":[0.02916666865348816,0.33796295523643494,0.9848959147930145,0.9694444239139557],"occluded":false},{"timestamp":12.48,"box":[0.2161458432674408,0.37962961196899414,0.6119791865348816,0.9694443941116333],"occluded":false},{"timestamp":12.74,"box":[0.03281250223517418,0.35185185074806213,0.9921875260770321,0.9712962806224823],"occluded":false},{"timestamp":13.0,"box":[0.03281250223517418,0.38333332538604736,0.9770834110677242,0.970370352268219],"occluded":false},{"timestamp":13.26,"box":[0.07083334028720856,0.39722222089767456,0.9562500268220901,0.9712963104248047],"occluded":false},{"timestamp":13.52,"box":[0.03229166939854622,0.39351850748062134,0.9552083797752857,0.9712962508201599],"occluded":false},{"timestamp":13.780000000000001,"box":[0.04218750074505806,0.4277777671813965,0.9911458976566792,0.970370352268219],"occluded":false},{"timestamp":14.040000000000001,"box":[0.10208334028720856,0.42129629850387573,0.9916667491197586,0.9712963104248047],"occluded":false},{"timestamp":14.3,"box":[0.0,0.4296296238899231,0.9776042103767395,0.970370352268219],"occluded":false},{"timestamp":14.56,"box":[0.0,0.36574074625968933,0.9505208730697632,0.9703703820705414],"occluded":false},{"timestamp":14.82,"box":[0.0,0.4009259045124054,0.9885417222976685,0.9712962806224823],"occluded":false},{"timestamp":15.08,"box":[0.0,0.4000000059604645,0.9953125715255737,0.9666666686534882],"occluded":false},{"timestamp":15.34,"box":[0.11614584177732468,0.37962961196899414,0.8255208656191826,0.9037036895751953],"occluded":false},{"timestamp":15.6,"box":[0.0,0.2481481432914734,0.9916667342185974,0.9564814567565918],"occluded":false},{"timestamp":15.860000000000001,"box":[0.2223958522081375,0.345370352268219,0.9911458641290665,0.8870370388031006],"occluded":false},{"timestamp":16.12,"box":[0.26875001192092896,0.3768518567085266,0.8520833849906921,0.7416666746139526],"occluded":false}],"scores":[0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295]},{"noun_phrase":"large boulder","object_id":8,"observations":[{"timestamp":1.82,"box":[0.4203125238418579,0.3861111104488373,0.5395833626389503,0.6361111104488373],"occluded":false},{"timestamp":2.08,"box":[0.14375001192092896,0.36574074625968933,0.5817708671092987,0.6129629611968994],"occluded":false},{"timestamp":2.34,"box":[0.17135417461395264,0.38055554032325745,0.5994791984558105,0.6287036836147308],"occluded":false},{"timestamp":2.6,"box":[0.19895833730697632,0.3444444239139557,0.6255208551883698,0.5879629403352737],"occluded":false},{"timestamp":2.86,"box":[0.19947917759418488,0.10925925523042679,0.9911458641290665,0.5851851776242256],"occluded":false},{"timestamp":3.12,"box":[0.24114584922790527,0.38333332538604736,0.6802083849906921,0.6361111104488373],"occluded":false},{"timestamp":3.38,"box":[0.24114584922790527,0.3861111104488373,0.6817708611488342,0.6444444358348846],"occluded":false},{"timestamp":3.64,"box":[0.13645833730697632,0.3583333194255829,0.6453125476837158,0.7574073970317841],"occluded":false},{"timestamp":3.9,"box":[0.07447917014360428,0.37870368361473083,0.6192708685994148,0.7814814448356628],"occluded":false},{"timestamp":4.16,"box":[0.10208334028720856,0.06666666269302368,0.9911458641290665,0.800000011920929],"occluded":false},{"timestamp":4.42,"box":[0.10208334028720856,0.06759259104728699,0.9911458641290665,0.7842592298984528],"occluded":false},{"timestamp":4.68,"box":[0.07447917014360428,0.3314814865589142,0.9911458566784859,0.8157407343387604],"occluded":false},{"timestamp":4.94,"box":[0.01875000074505806,0.317592591047287,0.9911458976566792,0.7962962985038757],"occluded":false},{"timestamp":5.2,"box":[0.03281250223517418,0.40740740299224854,0.6291666887700558,0.7842592597007751],"occluded":false},{"timestamp":5.46,"box":[0.03281250223517418,0.06666666269302368,0.9911458753049374,0.8111110925674438],"occluded":false},{"timestamp":5.72,"box":[0.07447917014360428,0.33796295523643494,0.6432291939854622,0.7972221970558167],"occluded":false},{"timestamp":5.98,"box":[0.03281250223517418,0.37962961196899414,0.6385417245328426,0.7972221970558167],"occluded":false},{"timestamp":6.24,"box":[0.03229166939854622,0.2611111104488373,0.6078125424683094,0.7620370090007782],"occluded":false},{"timestamp":6.5,"box":[0.10208334028720856,0.345370352268219,0.6348958760499954,0.7472221851348877],"occluded":false},{"timestamp":6.76,"box":[0.5291666984558105,0.5212962627410889,0.671354204416275,0.7722221910953522],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.10885417461395264,0.4000000059604645,0.675000011920929,0.7685185074806213],"occluded":false},{"timestamp":7.28,"box":[0.11510417610406876,0.5324074029922485,0.6682291999459267,0.8731481432914734],"occluded":false},{"timestamp":7.54,"box":[0.5119792222976685,0.6185185313224792,0.6578125655651093,0.8935185372829437],"occluded":false},{"timestamp":7.8,"box":[0.18541668355464935,0.5768518447875977,0.7119791954755783,0.8268518447875977],"occluded":false},{"timestamp":8.06,"box":[0.5911458730697632,0.575925886631012,0.7697917222976685,0.8240740299224854],"occluded":false},{"timestamp":8.32,"box":[0.19895833730697632,0.5185185074806213,0.7963541746139526,0.7620370239019394],"occluded":false},{"timestamp":8.58,"box":[0.14375001192092896,0.5185185074806213,0.7932292222976685,0.7814814746379852],"occluded":false},{"timestamp":8.84,"box":[0.12968750298023224,0.5046296119689941,0.7843750268220901,0.8129629492759705],"occluded":false},{"timestamp":9.1,"box":[0.07447917014360428,0.44907405972480774,0.7822917178273201,0.7962962687015533],"occluded":false},{"timestamp":9.36,"box":[0.03177083656191826,0.37962961196899414,0.8119791857898235,0.8101851642131805],"occluded":false},{"timestamp":9.620000000000001,"box":[0.5755208730697632,0.4564814865589142,0.99114590883255,0.8731481432914734],"occluded":false},{"timestamp":9.88,"box":[0.004687500186264515,0.4287036955356598,0.9770833970978856,0.8935185074806213],"occluded":false},{"timestamp":10.14,"box":[0.004166666883975267,0.42222222685813904,0.9078125399537385,0.8685185015201569],"occluded":false},{"timestamp":10.4,"box":[0.53125,0.4333333373069763,0.839062511920929,0.8990740776062012],"occluded":false},{"timestamp":10.66,"box":[0.5453125238418579,0.4166666567325592,0.8718750476837158,0.9074074029922485],"occluded":false},{"timestamp":10.92,"box":[0.5447916984558105,0.39629629254341125,0.8734375536441803,0.8888888657093048],"occluded":false},{"timestamp":11.18,"box":[0.5489583611488342,0.3731481432914734,0.8786458671092987,0.8648147881031036],"occluded":false},{"timestamp":11.44,"box":[0.06041666865348816,0.26851850748062134,0.866666704416275,0.855555534362793],"occluded":false},{"timestamp":11.700000000000001,"box":[0.5473958849906921,0.39537036418914795,0.8802084028720856,0.8888888657093048],"occluded":false},{"timestamp":11.96,"box":[0.534375011920929,0.41018518805503845,0.8630208671092987,0.9064814746379852],"occluded":false},{"timestamp":12.22,"box":[0.30677086114883423,0.4194444417953491,0.8541667461395264,0.9138888716697693],"occluded":false},{"timestamp":12.48,"box":[0.5260416865348816,0.4296296238899231,0.854166716337204,0.9259259104728699],"occluded":false},{"timestamp":12.74,"box":[0.534375011920929,0.43703702092170715,0.8630208671092987,0.9324073791503906],"occluded":false},{"timestamp":13.0,"box":[0.5348958373069763,0.43611109256744385,0.8614583611488342,0.930555522441864],"occluded":false},{"timestamp":13.26,"box":[0.515625,0.45185184478759766,0.8531250059604645,0.9527777433395386],"occluded":false},{"timestamp":13.52,"box":[0.5281250476837158,0.4851851761341095,0.8536458909511566,0.9712962806224823],"occluded":false},{"timestamp":13.780000000000001,"box":[0.534375011920929,0.4879629611968994,0.8604167103767395,0.9712962806224823],"occluded":false},{"timestamp":14.040000000000001,"box":[0.35364586114883423,0.4833333194255829,0.8609375357627869,0.970370352268219],"occluded":false},{"timestamp":14.3,"box":[0.06041666865348816,0.36574074625968933,0.8651042282581329,0.972222238779068],"occluded":false},{"timestamp":14.56,"box":[0.5375000238418579,0.4879629611968994,0.8656250536441803,0.970370352268219],"occluded":false},{"timestamp":14.82,"box":[0.53125,0.46481481194496155,0.8489583432674408,0.9388888776302338],"occluded":false},{"timestamp":15.08,"box":[0.0,0.45277777314186096,0.9994792342185974,0.9037036895751953],"occluded":false},{"timestamp":15.34,"box":[0.008333333767950535,0.4342592656612396,0.8265625322237611,0.8518518507480621],"occluded":false},{"timestamp":15.6,"box":[0.550000011920929,0.39907407760620117,0.8109375238418579,0.7805555462837219],"occluded":false},{"timestamp":15.860000000000001,"box":[0.5442708730697632,0.3768518567085266,0.7791667133569717,0.7166666686534882],"occluded":false},{"timestamp":16.12,"box":[0.07760416716337204,0.39537036418914795,0.744791679084301,0.6981481313705444],"occluded":false}],"scores":[0.8048780560493469,0.8048780560493469,0.8048780560493469,0.8048780560493469,0.8048780560493469,0.8048780560493469,0.8048780560493469,0.8048780560493469,0.8048780560493469,0.8048780560493469,0.8048780560493469,0.8048780560493469,0.8048780560493469,0.8048780560493469,0.8048780560493469,0.8048780560493469,0.8048780560493469,0.8048780560493469,0.8048780560493469,0.8048780560493469,0.8048780560493469,0.8048780560493469,0.8048780560493469,0.8048780560493469,0.8048780560493469,0.8048780560493469,0.8048780560493469,0.8048780560493469,0.8048780560493469,0.8048780560493469,0.8048780560493469,0.8048780560493469,0.8048780560493469,0.8048780560493469,0.8048780560493469,0.8048780560493469,0.8048780560493469,0.8048780560493469,0.8048780560493469,0.8048780560493469,0.8048780560493469,0.8048780560493469,0.8048780560493469,0.8048780560493469,0.8048780560493469,0.8048780560493469,0.8048780560493469,0.8048780560493469,0.8048780560493469,0.8048780560493469,0.8048780560493469,0.8048780560493469,0.8048780560493469,0.8048780560493469,0.8048780560493469,0.8048780560493469]},{"noun_phrase":"large boulder","object_id":9,"observations":[{"timestamp":2.08,"box":[0.4895833730697632,0.565740704536438,0.6098958775401115,0.7907406985759735],"occluded":false},{"timestamp":2.34,"box":[0.17135417461395264,0.40740740299224854,0.6338542103767395,0.8166666626930237],"occluded":false},{"timestamp":2.6,"box":[0.19895833730697632,0.42129629850387573,0.6645833551883698,0.7768518328666687],"occluded":false},{"timestamp":2.86,"box":[0.5489583611488342,0.5222222208976746,0.6781250387430191,0.7759259343147278],"occluded":false},{"timestamp":3.12,"box":[0.25468751788139343,0.44907405972480774,0.7250000536441803,0.8435184955596924],"occluded":false},{"timestamp":3.38,"box":[0.25468751788139343,0.46296295523643494,0.7281250357627869,0.8499999940395355],"occluded":false},{"timestamp":3.64,"box":[0.5531250238418579,0.6962962746620178,0.6937500238418579,0.9685184955596924],"occluded":false},{"timestamp":3.9,"box":[0.526562511920929,0.7157407402992249,0.673958346247673,0.970370352268219],"occluded":false},{"timestamp":4.16,"box":[0.004687500186264515,0.5564814805984497,0.7036458970978856,0.9694444239139557],"occluded":false},{"timestamp":4.42,"box":[0.5588542222976685,0.720370352268219,0.7145833969116211,0.970370352268219],"occluded":false},{"timestamp":4.68,"box":[0.5494791865348816,0.7555555701255798,0.7109375298023224,0.9703703820705414],"occluded":false},{"timestamp":4.94,"box":[0.5182291865348816,0.7361111044883728,0.677083358168602,0.970370352268219],"occluded":false},{"timestamp":5.2,"box":[0.15781250596046448,0.5324074029922485,0.6890625059604645,0.9694444239139557],"occluded":false},{"timestamp":5.46,"box":[0.53125,0.7509258985519409,0.7010416686534882,0.9703703373670578],"occluded":false},{"timestamp":5.72,"box":[0.534375011920929,0.7435185313224792,0.7026041895151138,0.9703703820705414],"occluded":false},{"timestamp":5.98,"box":[0.5250000357627869,0.7537037134170532,0.6984375417232513,0.9703703820705414],"occluded":false},{"timestamp":6.24,"box":[0.4921875298023224,0.7240740656852722,0.6687500327825546,0.9694444388151169],"occluded":false},{"timestamp":6.5,"box":[0.5281250476837158,0.6972222328186035,0.7031250596046448,0.9703703820705414],"occluded":false},{"timestamp":6.76,"box":[0.557812511920929,0.5148147940635681,0.7432291954755783,0.970370352268219],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.5625,0.690740704536438,0.7489583492279053,0.9712962508201599],"occluded":false},{"timestamp":7.28,"box":[0.550000011920929,0.800000011920929,0.7484375238418579,0.9712963104248047],"occluded":false},{"timestamp":7.54,"box":[0.43906253576278687,0.8064814805984497,0.741666704416275,0.9712962955236435],"occluded":false},{"timestamp":7.8,"box":[0.596875011920929,0.7259259223937988,0.8057291954755783,0.9703703671693802],"occluded":false},{"timestamp":8.06,"box":[0.24114584922790527,0.5601851940155029,0.8760417103767395,0.9703703820705414],"occluded":false},{"timestamp":8.32,"box":[0.19947917759418488,0.6175925731658936,0.9125000387430191,0.9712962806224823],"occluded":false},{"timestamp":8.58,"box":[0.6718750596046448,0.6425926089286804,0.9151042401790619,0.9712963104248047],"occluded":false},{"timestamp":8.84,"box":[0.39375001192092896,0.5324074029922485,0.9192708730697632,0.9712962806224823],"occluded":false},{"timestamp":9.1,"box":[0.4598958492279053,0.5564814805984497,0.9296875298023224,0.9712962806224823],"occluded":false},{"timestamp":9.36,"box":[0.4177083671092987,0.5287036895751953,0.991666704416275,0.9712962806224823],"occluded":false},{"timestamp":9.620000000000001,"box":[0.3895833492279053,0.5009258985519409,0.9994792342185974,0.9722221791744232],"occluded":false},{"timestamp":9.88,"box":[0.3369791805744171,0.4861111044883728,0.999479204416275,0.9712962806224823],"occluded":false},{"timestamp":10.14,"box":[0.3135416805744171,0.4166666567325592,0.999479204416275,0.9712962806224823],"occluded":false},{"timestamp":10.4,"box":[0.2927083373069763,0.4314814805984497,0.9994792342185974,0.9712963104248047],"occluded":false},{"timestamp":10.66,"box":[0.27916666865348816,0.3898147940635681,0.999479204416275,0.9712962508201599],"occluded":false},{"timestamp":10.92,"box":[0.27916666865348816,0.4037036895751953,0.999479204416275,0.9712962508201599],"occluded":false},{"timestamp":11.18,"box":[0.265625,0.4592592418193817,0.9994792342185974,0.9712962806224823],"occluded":false},{"timestamp":11.44,"box":[0.29218751192092896,0.3759259283542633,0.9994792342185974,0.9712962806224823],"occluded":false},{"timestamp":11.700000000000001,"box":[0.29322919249534607,0.345370352268219,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":11.96,"box":[0.27916666865348816,0.4037036895751953,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":12.22,"box":[0.2828125059604645,0.40740740299224854,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":12.48,"box":[0.29114586114883423,0.402777761220932,0.9994792342185974,0.9722221791744232],"occluded":false},{"timestamp":12.74,"box":[0.2645833492279053,0.4314814805984497,0.9994792342185974,0.9731481671333313],"occluded":false},{"timestamp":13.0,"box":[0.06041666865348816,0.28240740299224854,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":13.26,"box":[0.2645833492279053,0.42499998211860657,0.9994792342185974,0.9722221791744232],"occluded":false},{"timestamp":13.52,"box":[0.29218751192092896,0.4592592418193817,0.9994792342185974,0.9722221791744232],"occluded":false},{"timestamp":13.780000000000001,"box":[0.27916666865348816,0.47870370745658875,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":14.040000000000001,"box":[0.2786458432674408,0.4583333134651184,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":14.3,"box":[0.06041666865348816,0.36574074625968933,0.999479204416275,0.972222238779068],"occluded":false},{"timestamp":14.56,"box":[0.27916666865348816,0.4722222089767456,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":14.82,"box":[0.2786458432674408,0.49259257316589355,0.999479204416275,0.9712962806224823],"occluded":false},{"timestamp":15.08,"box":[0.24791668355464935,0.49259257316589355,0.9994792193174362,0.9712962806224823],"occluded":false},{"timestamp":15.34,"box":[0.31041666865348816,0.4407407343387604,0.999479204416275,0.9712962806224823],"occluded":false},{"timestamp":15.6,"box":[0.31666669249534607,0.3861111104488373,0.991666704416275,0.972222238779068],"occluded":false},{"timestamp":15.860000000000001,"box":[0.4489583671092987,0.3583333194255829,0.9494792520999908,0.9712962806224823],"occluded":false},{"timestamp":16.12,"box":[0.34427085518836975,0.3861111104488373,0.8838542401790619,0.9703703820705414],"occluded":false}],"scores":[0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933,0.6831682920455933]},{"noun_phrase":"large boulder","object_id":10,"observations":[{"timestamp":3.38,"box":[0.09687500447034836,0.25555554032325745,0.17916667461395264,0.36574072390794754],"occluded":false},{"timestamp":3.64,"box":[0.04843750223517418,0.32777777314186096,0.13385417684912682,0.38703703135252],"occluded":false},{"timestamp":3.9,"box":[0.0,0.33888888359069824,0.08906250447034836,0.4009259194135666],"occluded":false},{"timestamp":4.16,"box":[0.04010416939854622,0.3499999940395355,0.12708334252238274,0.41018517687916756],"occluded":false},{"timestamp":4.42,"box":[0.04114583507180214,0.35370370745658875,0.1276041753590107,0.40925926342606544],"occluded":false},{"timestamp":4.68,"box":[0.03437500074505806,0.3472222089767456,0.12343750521540642,0.3981481343507767],"occluded":false},{"timestamp":4.94,"box":[0.0,0.29907408356666565,0.0781250074505806,0.35370371118187904],"occluded":false},{"timestamp":5.2,"box":[0.0,0.2898148000240326,0.08645834028720856,0.3425925783813],"occluded":false},{"timestamp":5.46,"box":[0.0,0.3212963044643402,0.09531250596046448,0.3722222298383713],"occluded":false},{"timestamp":5.72,"box":[0.01302083395421505,0.3027777671813965,0.10312500409781933,0.35092591494321823],"occluded":false},{"timestamp":5.98,"box":[0.01770833507180214,0.2527777850627899,0.1093750037252903,0.304629635065794],"occluded":false},{"timestamp":6.24,"box":[0.0,0.15648147463798523,0.06406250596046448,0.21759258583188057],"occluded":false},{"timestamp":6.5,"box":[0.0,0.2342592477798462,0.07864584028720856,0.2925925813615322],"occluded":false},{"timestamp":6.76,"box":[0.0,0.32222220301628113,0.11093750596046448,0.3703703507781029],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.01510416716337204,0.3638888895511627,0.10364583879709244,0.40833333134651184],"occluded":false},{"timestamp":7.28,"box":[0.0,0.40648147463798523,0.0989583358168602,0.45277776941657066],"occluded":false},{"timestamp":7.54,"box":[0.0,0.38240739703178406,0.08125000447034836,0.4287036918103695],"occluded":false},{"timestamp":7.8,"box":[0.05000000447034836,0.46666666865348816,0.12552084028720856,0.5009259283542633],"occluded":false},{"timestamp":8.06,"box":[0.10364583879709244,0.5462962985038757,0.17656251043081284,0.5851851850748062],"occluded":false},{"timestamp":8.32,"box":[0.140625,0.5564814805984497,0.18958333507180214,0.5879629626870155],"occluded":false},{"timestamp":8.58,"box":[0.13593751192092896,0.5324074029922485,0.1901041828095913,0.5666666626930237],"occluded":false},{"timestamp":8.84,"box":[0.12291667610406876,0.5212962627410889,0.17552084475755692,0.5490740407258272],"occluded":false},{"timestamp":9.1,"box":[0.01250000111758709,0.4546296298503876,0.08229166828095913,0.49166666716337204],"occluded":false}],"scores":[0.7299578189849854,0.7299578189849854,0.7299578189849854,0.7299578189849854,0.7299578189849854,0.7299578189849854,0.7299578189849854,0.7299578189849854,0.7299578189849854,0.7299578189849854,0.7299578189849854,0.7299578189849854,0.7299578189849854,0.7299578189849854,0.7299578189849854,0.7299578189849854,0.7299578189849854,0.7299578189849854,0.7299578189849854,0.7299578189849854,0.7299578189849854,0.7299578189849854,0.7299578189849854]},{"noun_phrase":"large boulder","object_id":12,"observations":[{"timestamp":10.14,"box":[0.0,0.720370352268219,0.5281250476837158,0.9712962806224823],"occluded":false},{"timestamp":10.4,"box":[0.0,0.75,0.5088542103767395,0.9712962955236435],"occluded":false},{"timestamp":10.66,"box":[0.0,0.7490740418434143,0.5171875357627869,0.9722221791744232],"occluded":false},{"timestamp":10.92,"box":[0.0,0.7314814925193787,0.5302083492279053,0.9722222238779068],"occluded":false},{"timestamp":11.18,"box":[0.0,0.7064814567565918,0.5250000357627869,0.9712962806224823],"occluded":false},{"timestamp":11.44,"box":[0.0,0.7027777433395386,0.5213541984558105,0.9712962508201599],"occluded":false},{"timestamp":11.700000000000001,"box":[0.0,0.6472222208976746,0.5328125357627869,0.9722222089767456],"occluded":false},{"timestamp":11.96,"box":[0.0,0.7453703284263611,0.5104166865348816,0.9722221791744232],"occluded":false},{"timestamp":12.22,"box":[0.0,0.7518518567085266,0.49635419249534607,0.9722222238779068],"occluded":false},{"timestamp":12.48,"box":[0.0,0.760185182094574,0.48906251788139343,0.9722222089767456],"occluded":false},{"timestamp":12.74,"box":[0.0,0.7620370388031006,0.4932291805744171,0.9722222238779068],"occluded":false},{"timestamp":13.0,"box":[0.0,0.7592592239379883,0.4906250238418579,0.9722221791744232],"occluded":false},{"timestamp":13.26,"box":[0.0,0.7759259343147278,0.468229204416275,0.9722222238779068],"occluded":false},{"timestamp":13.52,"box":[0.0,0.8064814805984497,0.4322916865348816,0.9722222238779068],"occluded":false},{"timestamp":13.780000000000001,"box":[0.0,0.7981481552124023,0.4364583492279053,0.9722222238779068],"occluded":false},{"timestamp":14.040000000000001,"box":[0.0,0.7907407283782959,0.43802085518836975,0.9722222089767456],"occluded":false},{"timestamp":14.3,"box":[0.0,0.7990740537643433,0.43958336114883423,0.9722221940755844],"occluded":false},{"timestamp":14.56,"box":[0.0,0.8009259104728699,0.44322919845581055,0.9722222089767456],"occluded":false},{"timestamp":14.82,"box":[0.0,0.79537034034729,0.4854166805744171,0.9722221940755844],"occluded":false},{"timestamp":15.08,"box":[0.0,0.7842592597007751,0.5140625238418579,0.9722222238779068],"occluded":false},{"timestamp":15.34,"box":[0.0,0.7222222089767456,0.53125,0.9712962806224823],"occluded":false},{"timestamp":15.6,"box":[0.0,0.5796296000480652,0.5416666865348816,0.9712962508201599],"occluded":false},{"timestamp":15.860000000000001,"box":[0.0,0.5601851940155029,0.5364583730697632,0.9712963104248047],"occluded":false},{"timestamp":16.12,"box":[0.0078125,0.5592592358589172,0.5296875238418579,0.9694444239139557],"occluded":false}],"scores":[0.6604774594306946,0.6604774594306946,0.6604774594306946,0.6604774594306946,0.6604774594306946,0.6604774594306946,0.6604774594306946,0.6604774594306946,0.6604774594306946,0.6604774594306946,0.6604774594306946,0.6604774594306946,0.6604774594306946,0.6604774594306946,0.6604774594306946,0.6604774594306946,0.6604774594306946,0.6604774594306946,0.6604774594306946,0.6604774594306946,0.6604774594306946,0.6604774594306946,0.6604774594306946,0.6604774594306946]},{"noun_phrase":"large boulder","object_id":13,"observations":[{"timestamp":11.44,"box":[0.2979166805744171,0.1101851835846901,0.5270833522081375,0.3768518343567848],"occluded":false},{"timestamp":11.700000000000001,"box":[0.31041666865348816,0.14537036418914795,0.6020833551883698,0.40925925970077515],"occluded":false},{"timestamp":11.96,"box":[0.2958333492279053,0.15555554628372192,0.616666704416275,0.4296296238899231],"occluded":false},{"timestamp":12.22,"box":[0.04635417088866234,0.1648148149251938,0.6026042066514492,0.43796296417713165],"occluded":false},{"timestamp":12.48,"box":[0.2880208492279053,0.17314814031124115,0.5177083611488342,0.4407407194375992],"occluded":false},{"timestamp":12.74,"box":[0.29635417461395264,0.1787036955356598,0.5260416865348816,0.45277777314186096],"occluded":false},{"timestamp":13.0,"box":[0.06041666865348816,0.17777776718139648,0.5260416865348816,0.45185184478759766],"occluded":false},{"timestamp":13.26,"box":[0.10208334028720856,0.192592591047287,0.520833358168602,0.46759259700775146],"occluded":false},{"timestamp":13.52,"box":[0.28958335518836975,0.22592592239379883,0.5203125327825546,0.4990740716457367],"occluded":false},{"timestamp":13.780000000000001,"box":[0.06041666865348816,0.22499999403953552,0.5260416865348816,0.4990740716457367],"occluded":false},{"timestamp":14.040000000000001,"box":[0.12968750298023224,0.21851851046085358,0.5744792073965073,0.49166665971279144],"occluded":false},{"timestamp":14.3,"box":[0.06041666865348816,0.22592592239379883,0.6020833551883698,0.4990740716457367],"occluded":false},{"timestamp":14.56,"box":[0.07447917014360428,0.22962962090969086,0.6161458566784859,0.5018518418073654],"occluded":false},{"timestamp":14.82,"box":[0.07395833730697632,0.22685185074806213,0.6026042103767395,0.48055556416511536],"occluded":false},{"timestamp":15.08,"box":[0.15781250596046448,0.23333333432674408,0.5750000178813934,0.4712962955236435],"occluded":false},{"timestamp":15.34,"box":[0.1432291716337204,0.22407406568527222,0.5885417014360428,0.4425925761461258],"occluded":false},{"timestamp":15.6,"box":[0.18541668355464935,0.21203702688217163,0.6161458641290665,0.40925924479961395],"occluded":false},{"timestamp":15.860000000000001,"box":[0.3750000298023224,0.21296295523643494,0.5885417014360428,0.3916666507720947],"occluded":false},{"timestamp":16.12,"box":[0.3843750059604645,0.2490740716457367,0.5328125208616257,0.4009259194135666],"occluded":false}],"scores":[0.7730495929718018,0.7730495929718018,0.7730495929718018,0.7730495929718018,0.7730495929718018,0.7730495929718018,0.7730495929718018,0.7730495929718018,0.7730495929718018,0.7730495929718018,0.7730495929718018,0.7730495929718018,0.7730495929718018,0.7730495929718018,0.7730495929718018,0.7730495929718018,0.7730495929718018,0.7730495929718018,0.7730495929718018]},{"noun_phrase":"white sign","object_id":0,"observations":[{"timestamp":0.0,"box":[0.04531250149011612,0.47870370745658875,0.06458333507180214,0.5185185223817825],"occluded":false},{"timestamp":0.26,"box":[0.04791666939854622,0.5351851582527161,0.06562500447034836,0.5740740448236465],"occluded":false},{"timestamp":0.52,"box":[0.03697916865348816,0.4861111044883728,0.05781250260770321,0.5259259194135666],"occluded":false},{"timestamp":0.78,"box":[0.0390625037252903,0.4546296298503876,0.05885417200624943,0.49537036940455437],"occluded":false},{"timestamp":1.04,"box":[0.01770833507180214,0.4592592418193817,0.04114583693444729,0.5018518343567848],"occluded":false},{"timestamp":2.08,"box":[0.0015625001396983862,0.49074074625968933,0.027604168048128486,0.5398148186504841],"occluded":false},{"timestamp":2.34,"box":[0.02239583432674408,0.47870370745658875,0.04791666939854622,0.5259259305894375],"occluded":false},{"timestamp":2.6,"box":[0.05781250447034836,0.5018518567085266,0.08125000633299351,0.5462962985038757],"occluded":false},{"timestamp":2.86,"box":[0.06145833805203438,0.5009258985519409,0.09010417200624943,0.5462962687015533],"occluded":false},{"timestamp":3.12,"box":[0.11145833879709244,0.5120370388031006,0.13333334028720856,0.557407408952713],"occluded":false},{"timestamp":3.38,"box":[0.09843750298023224,0.5259259343147278,0.12291667051613331,0.5703703761100769],"occluded":false},{"timestamp":3.64,"box":[0.00885416753590107,0.6120370030403137,0.04218750260770321,0.6638888530433178],"occluded":false},{"timestamp":4.16,"box":[0.0,0.6324073672294617,0.004687500186264515,0.6499999593943357],"occluded":false}],"scores":[0.8552036285400391,0.8552036285400391,0.8552036285400391,0.8552036285400391,0.8552036285400391,0.8552036285400391,0.8552036285400391,0.8552036285400391,0.8552036285400391,0.8552036285400391,0.8552036285400391,0.8552036285400391,0.8552036285400391]},{"noun_phrase":"white sign","object_id":1,"observations":[{"timestamp":0.0,"box":[0.3161458373069763,0.6398147940635681,0.33385417237877846,0.6787036806344986],"occluded":false},{"timestamp":0.26,"box":[0.32239586114883423,0.6768518686294556,0.34062502905726433,0.7185185328125954],"occluded":false},{"timestamp":0.52,"box":[0.31562501192092896,0.6574074029922485,0.334895845502615,0.6990740671753883],"occluded":false},{"timestamp":0.78,"box":[0.32083335518836975,0.625,0.34010418877005577,0.6685185171663761],"occluded":false},{"timestamp":1.04,"box":[0.3114583492279053,0.6388888955116272,0.3322916831821203,0.6833333373069763],"occluded":false},{"timestamp":1.3,"box":[0.01875000074505806,0.595370352268219,0.28697918727993965,0.6740740537643433],"occluded":false},{"timestamp":1.56,"box":[0.24114584922790527,0.6064814925193787,0.26562501676380634,0.6537037156522274],"occluded":false},{"timestamp":1.82,"box":[0.27031251788139343,0.7138888835906982,0.2947916854172945,0.7638888843357563],"occluded":false},{"timestamp":2.08,"box":[0.31979167461395264,0.690740704536438,0.34166667610406876,0.7398147769272327],"occluded":false},{"timestamp":2.34,"box":[0.33229169249534607,0.700925886631012,0.35468752682209015,0.7490740343928337],"occluded":false},{"timestamp":2.6,"box":[0.36666667461395264,0.6814814805984497,0.3880208432674408,0.7296296283602715],"occluded":false},{"timestamp":2.86,"box":[0.07447917014360428,0.5120370388031006,0.39791668206453323,0.7277777791023254],"occluded":false},{"timestamp":3.12,"box":[0.4046875238418579,0.7175925970077515,0.42760419100522995,0.7685185223817825],"occluded":false},{"timestamp":3.38,"box":[0.40156251192092896,0.730555534362793,0.4250000137835741,0.782407384365797],"occluded":false},{"timestamp":3.64,"box":[0.01875000074505806,0.6231481432914734,0.3786458559334278,0.9120370149612427],"occluded":false},{"timestamp":3.9,"box":[0.31562501192092896,0.8870370388031006,0.3453125134110451,0.950925923883915],"occluded":false},{"timestamp":4.16,"box":[0.3343750238418579,0.8925925493240356,0.364583358168602,0.9583332911133766],"occluded":false},{"timestamp":4.42,"box":[0.004687500186264515,0.6361110806465149,0.36875001806765795,0.9518518149852753],"occluded":false},{"timestamp":4.68,"box":[0.004687500186264515,0.5537036657333374,0.35052086133509874,0.9731481075286865],"occluded":false},{"timestamp":4.94,"box":[0.10208334028720856,0.5120370388031006,0.31197918951511383,0.9564814865589142],"occluded":false},{"timestamp":5.2,"box":[0.28072917461395264,0.8675925731658936,0.31562500819563866,0.9416666477918625],"occluded":false},{"timestamp":5.46,"box":[0.2786458432674408,0.9009259343147278,0.315104179084301,0.9731481522321701],"occluded":false},{"timestamp":5.72,"box":[0.27916666865348816,0.8731481432914734,0.3161458373069763,0.9509259164333344],"occluded":false},{"timestamp":5.98,"box":[0.2588541805744171,0.8435184955596924,0.2979166842997074,0.92314812541008],"occluded":false},{"timestamp":6.24,"box":[0.21406251192092896,0.7990740537643433,0.2567708492279053,0.8796296045184135],"occluded":false},{"timestamp":6.5,"box":[0.2526041865348816,0.8212962746620178,0.29322918877005577,0.9046296030282974],"occluded":false},{"timestamp":6.76,"box":[0.2796875238418579,0.8666666746139526,0.31927086040377617,0.9518518596887589],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.27916666865348816,0.8907407522201538,0.31822917237877846,0.9731481596827507],"occluded":false}],"scores":[0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258]},{"noun_phrase":"white sign","object_id":2,"observations":[{"timestamp":0.0,"box":[0.3343750238418579,0.5537036657333374,0.38854169473052025,0.6564814448356628],"occluded":false},{"timestamp":0.26,"box":[0.34010419249534607,0.5888888835906982,0.39479169622063637,0.6916666626930237],"occluded":false},{"timestamp":0.52,"box":[0.3348958492279053,0.5694444179534912,0.3916666842997074,0.6851851567626],"occluded":false},{"timestamp":0.78,"box":[0.3395833373069763,0.5379629731178284,0.39635417237877846,0.6574074178934097],"occluded":false},{"timestamp":1.04,"box":[0.33125001192092896,0.5490740537643433,0.3916666805744171,0.6712962761521339],"occluded":false},{"timestamp":1.3,"box":[0.28697916865348816,0.5361111164093018,0.3526041731238365,0.6537037044763565],"occluded":false},{"timestamp":1.56,"box":[0.26510417461395264,0.5120370388031006,0.3322916775941849,0.6435185223817825],"occluded":false},{"timestamp":1.82,"box":[0.29375001788139343,0.614814817905426,0.36093752086162567,0.7407407462596893],"occluded":false},{"timestamp":2.08,"box":[0.3411458432674408,0.5944444537162781,0.4062500149011612,0.7268518656492233],"occluded":false},{"timestamp":2.34,"box":[0.3541666865348816,0.6018518209457397,0.4203125238418579,0.7407407015562057],"occluded":false},{"timestamp":2.6,"box":[0.3864583671092987,0.5805555582046509,0.4500000402331352,0.7018518522381783],"occluded":false},{"timestamp":2.86,"box":[0.39531251788139343,0.5777777433395386,0.46041668951511383,0.6999999657273293],"occluded":false},{"timestamp":3.12,"box":[0.4270833432674408,0.614814817905426,0.4932291805744171,0.7453703731298447],"occluded":false},{"timestamp":3.38,"box":[0.4244791865348816,0.6240740418434143,0.49375002086162567,0.7564814537763596],"occluded":false},{"timestamp":3.64,"box":[0.3786458671092987,0.7342592477798462,0.45364587008953094,0.8935185074806213],"occluded":false},{"timestamp":3.9,"box":[0.3463541865348816,0.6129629611968994,0.48854169249534607,0.9148148000240326],"occluded":false},{"timestamp":4.16,"box":[0.3656250238418579,0.7694444060325623,0.4463541954755783,0.9268518090248108],"occluded":false},{"timestamp":4.42,"box":[0.004687500186264515,0.6370370388031006,0.4515625359490514,0.9185185134410858],"occluded":false},{"timestamp":4.68,"box":[0.004687500186264515,0.5537036657333374,0.4395833434537053,0.9490740299224854],"occluded":false},{"timestamp":4.94,"box":[0.3140625059604645,0.7509258985519409,0.4072916805744171,0.9249999672174454],"occluded":false},{"timestamp":5.2,"box":[0.3177083432674408,0.7351851463317871,0.41197918355464935,0.9120370000600815],"occluded":false},{"timestamp":5.46,"box":[0.3177083432674408,0.7648147940635681,0.4151041805744171,0.9490740448236465],"occluded":false},{"timestamp":5.72,"box":[0.3177083432674408,0.7407407164573669,0.41718751192092896,0.9240740537643433],"occluded":false},{"timestamp":5.98,"box":[0.30156251788139343,0.7138888835906982,0.40781252086162567,0.9074074029922485],"occluded":false},{"timestamp":6.24,"box":[0.26093751192092896,0.6685184836387634,0.37291668355464935,0.866666629910469],"occluded":false},{"timestamp":6.5,"box":[0.29635417461395264,0.6833333373069763,0.4052083492279053,0.8777777850627899],"occluded":false},{"timestamp":6.76,"box":[0.3239583373069763,0.7240740656852722,0.42916667461395264,0.9166666567325592],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.3229166865348816,0.7379629611968994,0.43177086114883423,0.9351851791143417],"occluded":false},{"timestamp":7.28,"box":[0.17135417461395264,0.8138888478279114,0.4125000238418579,0.9990740269422531],"occluded":false},{"timestamp":7.54,"box":[0.18541668355464935,0.6712962985038757,0.45625002682209015,0.9990740716457367],"occluded":false},{"timestamp":7.8,"box":[0.24114584922790527,0.8083333373069763,0.45364585518836975,0.9990740716457367],"occluded":false},{"timestamp":8.06,"box":[0.2828125059604645,0.8388888835906982,0.518750011920929,0.9990740716457367],"occluded":false},{"timestamp":8.32,"box":[0.4255208671092987,0.8064814805984497,0.5322917029261589,0.9722222238779068],"occluded":false},{"timestamp":8.58,"box":[0.4130208492279053,0.7990740537643433,0.5744791924953461,0.9990740567445755],"occluded":false},{"timestamp":8.84,"box":[0.06041666865348816,0.8166666626930237,0.5088541805744171,0.9990740716457367],"occluded":false},{"timestamp":9.1,"box":[0.32447919249534607,0.7944444417953491,0.476041704416275,0.9990740716457367],"occluded":false},{"timestamp":9.36,"box":[0.30000001192092896,0.815740704536438,0.450520858168602,0.9990740418434143],"occluded":false},{"timestamp":9.620000000000001,"box":[0.24843750894069672,0.8583332896232605,0.41979168355464935,0.9990740269422531],"occluded":false},{"timestamp":9.88,"box":[0.20156250894069672,0.8888888955116272,0.39270834624767303,0.9722222238779068],"occluded":false},{"timestamp":10.14,"box":[0.14166668057441711,0.8518518209457397,0.35104168951511383,0.9722221866250038],"occluded":false},{"timestamp":10.4,"box":[0.09479167312383652,0.8787037134170532,0.31458335369825363,0.9722222313284874],"occluded":false},{"timestamp":10.66,"box":[0.08958333730697632,0.8796296119689941,0.3177083432674408,0.972222201526165],"occluded":false},{"timestamp":10.92,"box":[0.08645834028720856,0.8638888597488403,0.31718751788139343,0.9722221940755844],"occluded":false},{"timestamp":11.18,"box":[0.09218750149011612,0.8388888835906982,0.3244791850447655,0.9722222089767456],"occluded":false},{"timestamp":11.44,"box":[0.08229167014360428,0.8361110687255859,0.3161458447575569,0.9722221791744232],"occluded":false},{"timestamp":11.700000000000001,"box":[0.09687500447034836,0.8694444298744202,0.32500001043081284,0.9722222089767456],"occluded":false},{"timestamp":11.96,"box":[0.07864584028720856,0.8787037134170532,0.30885419249534607,0.9722222313284874],"occluded":false},{"timestamp":12.22,"box":[0.06927083432674408,0.8879629373550415,0.2994791865348816,0.9722221940755844],"occluded":false},{"timestamp":12.48,"box":[0.06875000149011612,0.895370364189148,0.2984375134110451,0.9722222164273262],"occluded":false},{"timestamp":12.74,"box":[0.07864584028720856,0.8972222208976746,0.3072916865348816,0.9731481447815895],"occluded":false},{"timestamp":13.0,"box":[0.07708333432674408,0.8944444060325623,0.30520834028720856,0.9722221791744232],"occluded":false},{"timestamp":13.26,"box":[0.06822916865348816,0.9111111164093018,0.2776041775941849,0.9722222276031971],"occluded":false},{"timestamp":13.52,"box":[0.06979166716337204,0.9435185194015503,0.17239584028720856,0.9722222220152617],"occluded":false},{"timestamp":13.780000000000001,"box":[0.07604166865348816,0.9342592358589172,0.1979166716337204,0.9722221978008747],"occluded":false},{"timestamp":14.040000000000001,"box":[0.07604166865348816,0.9259259104728699,0.2213541716337204,0.9722222052514553],"occluded":false},{"timestamp":14.3,"box":[0.07916667312383652,0.9351851940155029,0.19687501341104507,0.9722222313284874],"occluded":false},{"timestamp":14.56,"box":[0.08437500149011612,0.9351851940155029,0.213541679084301,0.9722222313284874],"occluded":false},{"timestamp":14.82,"box":[0.10208334028720856,0.9268518090248108,0.28333334624767303,0.9722221791744232],"occluded":false},{"timestamp":15.08,"box":[0.12343750894069672,0.9101851582527161,0.33020834624767303,0.9722221940755844],"occluded":false},{"timestamp":15.34,"box":[0.16197916865348816,0.8370370268821716,0.3593750149011612,0.9722222089767456],"occluded":false},{"timestamp":15.6,"box":[0.20156250894069672,0.7777777910232544,0.38593751192092896,0.972222238779068],"occluded":false}],"scores":[0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708]},{"noun_phrase":"white sign","object_id":3,"observations":[{"timestamp":0.0,"box":[0.3541666865348816,0.6499999761581421,0.36406252067536116,0.674074050039053],"occluded":false},{"timestamp":0.26,"box":[0.3609375059604645,0.6870370507240295,0.37187500670552254,0.7101851981133223],"occluded":false},{"timestamp":0.52,"box":[0.35468751192092896,0.6694444417953491,0.366145845502615,0.69351851567626],"occluded":false},{"timestamp":0.78,"box":[0.35989585518836975,0.6379629373550415,0.37135418877005577,0.6648147888481617],"occluded":false},{"timestamp":1.04,"box":[0.3515625298023224,0.6509259343147278,0.36562503036111593,0.6796296369284391],"occluded":false},{"timestamp":1.3,"box":[0.30781251192092896,0.6435185074806213,0.32239584624767303,0.6722222100943327],"occluded":false},{"timestamp":1.56,"box":[0.28541669249534607,0.6259258985519409,0.30000002682209015,0.6546296011656523],"occluded":false},{"timestamp":1.82,"box":[0.31562501192092896,0.7287036776542664,0.33125001192092896,0.7583333067595959],"occluded":false},{"timestamp":2.08,"box":[0.3630208671092987,0.7074074149131775,0.37708336766809225,0.7361111175268888],"occluded":false},{"timestamp":2.34,"box":[0.3765625059604645,0.7175925970077515,0.390625006519258,0.7481481526046991],"occluded":false},{"timestamp":2.6,"box":[0.40937501192092896,0.6944444179534912,0.42395834624767303,0.7231481205672026],"occluded":false},{"timestamp":2.86,"box":[0.41875001788139343,0.6916666626930237,0.43489585258066654,0.7212962917983532],"occluded":false},{"timestamp":3.12,"box":[0.45000001788139343,0.7342592477798462,0.464062518440187,0.766666654497385],"occluded":false},{"timestamp":3.38,"box":[0.44843751192092896,0.7462962865829468,0.46406251192092896,0.7777777686715126],"occluded":false},{"timestamp":3.64,"box":[0.40312501788139343,0.8703703284263611,0.4197916854172945,0.9074073657393456],"occluded":false},{"timestamp":3.9,"box":[0.37135419249534607,0.9046295881271362,0.3890625275671482,0.9435184746980667],"occluded":false},{"timestamp":4.16,"box":[0.3916666805744171,0.9138888716697693,0.40885418094694614,0.9537036865949631],"occluded":false},{"timestamp":4.42,"box":[0.3968750238418579,0.9055555462837219,0.415104191750288,0.9444444328546524],"occluded":false},{"timestamp":4.68,"box":[0.37916669249534607,0.9287036657333374,0.39791669324040413,0.9712962582707405],"occluded":false},{"timestamp":4.94,"box":[0.34166669845581055,0.9092592597007751,0.3625000324100256,0.9527777768671513],"occluded":false},{"timestamp":5.2,"box":[0.34583336114883423,0.8944444060325623,0.3666666951030493,0.9388888478279114],"occluded":false},{"timestamp":5.46,"box":[0.3473958373069763,0.9287036657333374,0.3677083384245634,0.9722221828997135],"occluded":false},{"timestamp":5.72,"box":[0.34791669249534607,0.9055555462837219,0.3687500264495611,0.949999988079071],"occluded":false},{"timestamp":5.98,"box":[0.3291666805744171,0.8824073672294617,0.3531250152736902,0.9296295903623104],"occluded":false},{"timestamp":6.24,"box":[0.28958335518836975,0.8398147821426392,0.31354168988764286,0.8870370052754879],"occluded":false},{"timestamp":6.5,"box":[0.32604169845581055,0.854629635810852,0.3536458667367697,0.9055555611848831],"occluded":false},{"timestamp":6.76,"box":[0.3552083373069763,0.8981481194496155,0.3796875048428774,0.9462962672114372],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.3583333492279053,0.9166666269302368,0.3822916839271784,0.9694444052875042],"occluded":false}],"scores":[0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458]},{"noun_phrase":"white sign","object_id":4,"observations":[{"timestamp":0.0,"box":[0.2864583432674408,0.6324073672294617,0.2989583443850279,0.6601851452142],"occluded":false},{"timestamp":0.26,"box":[0.2916666865348816,0.6722221970558167,0.3052083542570472,0.7018518261611462],"occluded":false},{"timestamp":0.52,"box":[0.2848958373069763,0.6490740776062012,0.2984375050291419,0.6787037067115307],"occluded":false},{"timestamp":0.78,"box":[0.2890625298023224,0.6185185313224792,0.30312503036111593,0.6481481604278088],"occluded":false},{"timestamp":1.04,"box":[0.2796875238418579,0.6314814686775208,0.2953125238418579,0.6620370242744684],"occluded":false},{"timestamp":1.3,"box":[0.2302083522081375,0.6175925731658936,0.24635418690741062,0.6499999798834324],"occluded":false},{"timestamp":1.56,"box":[0.20468750596046448,0.5925925970077515,0.22239584103226662,0.6287037059664726],"occluded":false},{"timestamp":1.82,"box":[0.2343750149011612,0.7046296000480652,0.25312501564621925,0.7398147843778133],"occluded":false},{"timestamp":2.08,"box":[0.2848958373069763,0.6824073791503906,0.30104167200624943,0.7175925634801388],"occluded":false},{"timestamp":2.34,"box":[0.2979166805744171,0.6888888478279114,0.3140625152736902,0.7240740321576595],"occluded":false},{"timestamp":2.6,"box":[0.33125001192092896,0.6740740537643433,0.34791667945683,0.7074073851108551],"occluded":false},{"timestamp":2.86,"box":[0.33906251192092896,0.6722221970558167,0.35729167982935905,0.7074073813855648],"occluded":false},{"timestamp":3.12,"box":[0.36927086114883423,0.7074074149131775,0.38541669584810734,0.7425925992429256],"occluded":false},{"timestamp":3.38,"box":[0.3656250238418579,0.720370352268219,0.38281252421438694,0.7564814612269402],"occluded":false},{"timestamp":3.64,"box":[0.31197917461395264,0.8398147821426392,0.33125000819563866,0.8833332993090153],"occluded":false},{"timestamp":3.9,"box":[0.2723958492279053,0.8759258985519409,0.2942708507180214,0.9212962687015533],"occluded":false},{"timestamp":4.16,"box":[0.29218751192092896,0.8777777552604675,0.313020845875144,0.92314812541008],"occluded":false},{"timestamp":4.42,"box":[0.2968750298023224,0.8731481432914734,0.3171875309199095,0.9185185134410858],"occluded":false},{"timestamp":4.68,"box":[0.2734375,0.8879629373550415,0.2973958346992731,0.9370370097458363],"occluded":false},{"timestamp":4.94,"box":[0.22968751192092896,0.8666666746139526,0.256250012665987,0.9175925999879837],"occluded":false},{"timestamp":5.2,"box":[0.23333334922790527,0.8499999642372131,0.25833335146307945,0.9009258896112442],"occluded":false},{"timestamp":5.46,"box":[0.2302083522081375,0.8842592239379883,0.2572916876524687,0.9379629269242287],"occluded":false},{"timestamp":5.72,"box":[0.2302083522081375,0.854629635810852,0.25781252048909664,0.9083333387970924],"occluded":false},{"timestamp":5.98,"box":[0.21145834028720856,0.8175925612449646,0.24114584177732468,0.8740740418434143],"occluded":false},{"timestamp":6.24,"box":[0.16302084922790527,0.7712962627410889,0.19583335146307945,0.8277777433395386],"occluded":false},{"timestamp":6.5,"box":[0.20000001788139343,0.8009259104728699,0.23229168727993965,0.8592592440545559],"occluded":false},{"timestamp":6.76,"box":[0.22760418057441711,0.8509259223937988,0.25625001452863216,0.9074074029922485],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.2239583432674408,0.8805555105209351,0.254687512293458,0.9370369911193848],"occluded":false},{"timestamp":7.28,"box":[0.18697917461395264,0.9574074149131775,0.2156250085681677,0.9722222294658422],"occluded":false},{"timestamp":7.54,"box":[0.16302084922790527,0.9537037014961243,0.19114585034549236,0.9722222201526165],"occluded":false}],"scores":[0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258]}],"tracklets":[{"noun_phrase":"large boulder","object_id":8,"presentation_id":1,"observations":[{"timestamp":1.82,"box":[0.4203125238418579,0.3861111104488373,0.5395833626389503,0.6361111104488373],"occluded":false},{"timestamp":2.08,"box":[0.14375001192092896,0.36574074625968933,0.5817708671092987,0.6129629611968994],"occluded":false},{"timestamp":2.34,"box":[0.17135417461395264,0.38055554032325745,0.5994791984558105,0.6287036836147308],"occluded":false},{"timestamp":2.6,"box":[0.19895833730697632,0.3444444239139557,0.6255208551883698,0.5879629403352737],"occluded":false},{"timestamp":2.86,"box":[0.19947917759418488,0.10925925523042679,0.9911458641290665,0.5851851776242256],"occluded":false},{"timestamp":3.12,"box":[0.24114584922790527,0.38333332538604736,0.6802083849906921,0.6361111104488373],"occluded":false},{"timestamp":3.38,"box":[0.24114584922790527,0.3861111104488373,0.6817708611488342,0.6444444358348846],"occluded":false},{"timestamp":3.64,"box":[0.13645833730697632,0.3583333194255829,0.6453125476837158,0.7574073970317841],"occluded":false},{"timestamp":3.9,"box":[0.07447917014360428,0.37870368361473083,0.6192708685994148,0.7814814448356628],"occluded":false},{"timestamp":4.16,"box":[0.10208334028720856,0.06666666269302368,0.9911458641290665,0.800000011920929],"occluded":false},{"timestamp":4.42,"box":[0.10208334028720856,0.06759259104728699,0.9911458641290665,0.7842592298984528],"occluded":false},{"timestamp":4.68,"box":[0.07447917014360428,0.3314814865589142,0.9911458566784859,0.8157407343387604],"occluded":false},{"timestamp":4.94,"box":[0.01875000074505806,0.317592591047287,0.9911458976566792,0.7962962985038757],"occluded":false},{"timestamp":5.2,"box":[0.03281250223517418,0.40740740299224854,0.6291666887700558,0.7842592597007751],"occluded":false},{"timestamp":5.46,"box":[0.03281250223517418,0.06666666269302368,0.9911458753049374,0.8111110925674438],"occluded":false},{"timestamp":5.72,"box":[0.07447917014360428,0.33796295523643494,0.6432291939854622,0.7972221970558167],"occluded":false},{"timestamp":5.98,"box":[0.03281250223517418,0.37962961196899414,0.6385417245328426,0.7972221970558167],"occluded":false},{"timestamp":6.24,"box":[0.03229166939854622,0.2611111104488373,0.6078125424683094,0.7620370090007782],"occluded":false},{"timestamp":6.5,"box":[0.10208334028720856,0.345370352268219,0.6348958760499954,0.7472221851348877],"occluded":false},{"timestamp":6.76,"box":[0.5291666984558105,0.5212962627410889,0.671354204416275,0.7722221910953522],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.10885417461395264,0.4000000059604645,0.675000011920929,0.7685185074806213],"occluded":false},{"timestamp":7.28,"box":[0.11510417610406876,0.5324074029922485,0.6682291999459267,0.8731481432914734],"occluded":false},{"timestamp":7.54,"box":[0.5119792222976685,0.6185185313224792,0.6578125655651093,0.8935185372829437],"occluded":false},{"timestamp":7.8,"box":[0.18541668355464935,0.5768518447875977,0.7119791954755783,0.8268518447875977],"occluded":false},{"timestamp":8.06,"box":[0.5911458730697632,0.575925886631012,0.7697917222976685,0.8240740299224854],"occluded":false},{"timestamp":8.32,"box":[0.19895833730697632,0.5185185074806213,0.7963541746139526,0.7620370239019394],"occluded":false},{"timestamp":8.58,"box":[0.14375001192092896,0.5185185074806213,0.7932292222976685,0.7814814746379852],"occluded":false},{"timestamp":8.84,"box":[0.12968750298023224,0.5046296119689941,0.7843750268220901,0.8129629492759705],"occluded":false},{"timestamp":9.1,"box":[0.07447917014360428,0.44907405972480774,0.7822917178273201,0.7962962687015533],"occluded":false},{"timestamp":9.36,"box":[0.03177083656191826,0.37962961196899414,0.8119791857898235,0.8101851642131805],"occluded":false},{"timestamp":9.620000000000001,"box":[0.5755208730697632,0.4564814865589142,0.99114590883255,0.8731481432914734],"occluded":false},{"timestamp":9.88,"box":[0.004687500186264515,0.4287036955356598,0.9770833970978856,0.8935185074806213],"occluded":false},{"timestamp":10.14,"box":[0.004166666883975267,0.42222222685813904,0.9078125399537385,0.8685185015201569],"occluded":false},{"timestamp":10.4,"box":[0.53125,0.4333333373069763,0.839062511920929,0.8990740776062012],"occluded":false},{"timestamp":10.66,"box":[0.5453125238418579,0.4166666567325592,0.8718750476837158,0.9074074029922485],"occluded":false},{"timestamp":10.92,"box":[0.5447916984558105,0.39629629254341125,0.8734375536441803,0.8888888657093048],"occluded":false},{"timestamp":11.18,"box":[0.5489583611488342,0.3731481432914734,0.8786458671092987,0.8648147881031036],"occluded":false},{"timestamp":11.44,"box":[0.06041666865348816,0.26851850748062134,0.866666704416275,0.855555534362793],"occluded":false},{"timestamp":11.700000000000001,"box":[0.5473958849906921,0.39537036418914795,0.8802084028720856,0.8888888657093048],"occluded":false},{"timestamp":11.96,"box":[0.534375011920929,0.41018518805503845,0.8630208671092987,0.9064814746379852],"occluded":false},{"timestamp":12.22,"box":[0.30677086114883423,0.4194444417953491,0.8541667461395264,0.9138888716697693],"occluded":false},{"timestamp":12.48,"box":[0.5260416865348816,0.4296296238899231,0.854166716337204,0.9259259104728699],"occluded":false},{"timestamp":12.74,"box":[0.534375011920929,0.43703702092170715,0.8630208671092987,0.9324073791503906],"occluded":false},{"timestamp":13.0,"box":[0.5348958373069763,0.43611109256744385,0.8614583611488342,0.930555522441864],"occluded":false},{"timestamp":13.26,"box":[0.515625,0.45185184478759766,0.8531250059604645,0.9527777433395386],"occluded":false},{"timestamp":13.52,"box":[0.5281250476837158,0.4851851761341095,0.8536458909511566,0.9712962806224823],"occluded":false},{"timestamp":13.780000000000001,"box":[0.534375011920929,0.4879629611968994,0.8604167103767395,0.9712962806224823],"occluded":false},{"timestamp":14.040000000000001,"box":[0.35364586114883423,0.4833333194255829,0.8609375357627869,0.970370352268219],"occluded":false},{"timestamp":14.3,"box":[0.06041666865348816,0.36574074625968933,0.8651042282581329,0.972222238779068],"occluded":false},{"timestamp":14.56,"box":[0.5375000238418579,0.4879629611968994,0.8656250536441803,0.970370352268219],"occluded":false},{"timestamp":14.82,"box":[0.53125,0.46481481194496155,0.8489583432674408,0.9388888776302338],"occluded":false},{"timestamp":15.08,"box":[0.0,0.45277777314186096,0.9994792342185974,0.9037036895751953],"occluded":false},{"timestamp":15.34,"box":[0.008333333767950535,0.4342592656612396,0.8265625322237611,0.8518518507480621],"occluded":false},{"timestamp":15.6,"box":[0.550000011920929,0.39907407760620117,0.8109375238418579,0.7805555462837219],"occluded":false},{"timestamp":15.860000000000001,"box":[0.5442708730697632,0.3768518567085266,0.7791667133569717,0.7166666686534882],"occluded":false},{"timestamp":16.12,"box":[0.07760416716337204,0.39537036418914795,0.744791679084301,0.6981481313705444],"occluded":false}],"mean_score":0.8048780560493469},{"noun_phrase":"large boulder","object_id":4,"presentation_id":2,"observations":[{"timestamp":0.0,"box":[0.33645835518836975,0.18518517911434174,0.4802083671092987,0.31018517911434174],"occluded":false},{"timestamp":0.26,"box":[0.33385419845581055,0.2231481373310089,0.47968754172325134,0.34166665375232697],"occluded":false},{"timestamp":0.52,"box":[0.3411458432674408,0.20555555820465088,0.4869791865348816,0.3305555582046509],"occluded":false},{"timestamp":0.78,"box":[0.34427085518836975,0.17222221195697784,0.49166668951511383,0.3009259104728699],"occluded":false},{"timestamp":1.04,"box":[0.3395833373069763,0.1814814805984497,0.48750001192092896,0.3157407343387604],"occluded":false},{"timestamp":1.3,"box":[0.3083333373069763,0.17037037014961243,0.4598958492279053,0.31481480598449707],"occluded":false},{"timestamp":1.56,"box":[0.29114586114883423,0.14629629254341125,0.4453125298023224,0.3018518388271332],"occluded":false},{"timestamp":1.82,"box":[0.3151041865348816,0.25462961196899414,0.536458358168602,0.5601851642131805],"occluded":false},{"timestamp":2.08,"box":[0.3453125059604645,0.23888888955116272,0.502083346247673,0.3740740716457367],"occluded":false},{"timestamp":2.34,"box":[0.37552085518836975,0.2490740716457367,0.520833358168602,0.38055555522441864],"occluded":false},{"timestamp":2.6,"box":[0.3916666805744171,0.2222222238779068,0.5390625149011612,0.3462962955236435],"occluded":false},{"timestamp":2.86,"box":[0.31041666865348816,0.21944443881511688,0.5468750149011612,0.33888888359069824],"occluded":false},{"timestamp":3.12,"box":[0.008333333767950535,0.17129629850387573,0.7552083814516664,0.5185185074806213],"occluded":false},{"timestamp":3.38,"box":[0.4427083432674408,0.27037036418914795,0.5901041775941849,0.39537036418914795],"occluded":false},{"timestamp":3.64,"box":[0.4072916805744171,0.3759259283542633,0.6890625357627869,0.7518518567085266],"occluded":false},{"timestamp":3.9,"box":[0.38072919845581055,0.39907407760620117,0.5744792073965073,0.7824074029922485],"occluded":false},{"timestamp":4.16,"box":[0.40781253576278687,0.06759259104728699,0.99114590883255,0.5388888716697693],"occluded":false},{"timestamp":4.42,"box":[0.41197919845581055,0.4009259045124054,0.7000000476837158,0.7268518209457397],"occluded":false},{"timestamp":4.68,"box":[0.06041666865348816,0.3314814865589142,0.9911458790302277,0.5509259253740311],"occluded":false},{"timestamp":4.94,"box":[0.38229167461395264,0.4000000059604645,0.5312500149011612,0.5333333313465118],"occluded":false},{"timestamp":5.2,"box":[0.39010417461395264,0.3888888955116272,0.5401041805744171,0.5250000059604645],"occluded":false},{"timestamp":5.46,"box":[0.39635419845581055,0.4120370149612427,0.5604167133569717,0.9490740299224854],"occluded":false},{"timestamp":5.72,"box":[0.40052086114883423,0.4009259045124054,0.5604166984558105,0.8101851642131805],"occluded":false},{"timestamp":5.98,"box":[0.4052083492279053,0.39259257912635803,0.5541666895151138,0.5388888716697693],"occluded":false},{"timestamp":6.24,"box":[0.37812501192092896,0.35185185074806213,0.528645858168602,0.5074073970317841],"occluded":false},{"timestamp":6.5,"box":[0.3958333432674408,0.35462963581085205,0.5458333492279053,0.49537037312984467],"occluded":false},{"timestamp":6.76,"box":[0.41979169845581055,0.3861111104488373,0.5729167014360428,0.5175925940275192],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.4166666865348816,0.39444443583488464,0.5708333551883698,0.5212962925434113],"occluded":false},{"timestamp":7.28,"box":[0.41093751788139343,0.47685185074806213,0.5656250268220901,0.614814817905426],"occluded":false},{"timestamp":7.54,"box":[0.40000003576278687,0.47592592239379883,0.5562500506639481,0.6157407313585281],"occluded":false},{"timestamp":7.8,"box":[0.43697917461395264,0.4564814865589142,0.5927083492279053,0.5814814865589142],"occluded":false},{"timestamp":8.06,"box":[0.37968751788139343,0.4722222089767456,0.632291704416275,0.585185170173645],"occluded":false},{"timestamp":8.32,"box":[0.008333333767950535,0.4314814805984497,0.6510416949167848,0.5601851791143417],"occluded":false},{"timestamp":8.58,"box":[0.26875001192092896,0.39351850748062134,0.9354166984558105,0.5481481403112411],"occluded":false},{"timestamp":8.84,"box":[0.12968750298023224,0.44999998807907104,0.6453125029802322,0.5666666552424431],"occluded":false},{"timestamp":9.1,"box":[0.4489583671092987,0.38148146867752075,0.632291704416275,0.5194444358348846],"occluded":false},{"timestamp":9.36,"box":[0.03281250223517418,0.3203703761100769,0.6375000141561031,0.47870370745658875],"occluded":false},{"timestamp":9.620000000000001,"box":[0.3906250298023224,0.28703704476356506,0.9494792520999908,0.8518518507480621],"occluded":false},{"timestamp":9.88,"box":[0.3687500059604645,0.2620370388031006,0.6500000059604645,0.47870370745658875],"occluded":false},{"timestamp":10.14,"box":[0.04635417088866234,0.20000000298023224,0.9770833812654018,0.7129629403352737],"occluded":false},{"timestamp":10.4,"box":[0.14375001192092896,0.1944444477558136,0.6171875298023224,0.45277777314186096],"occluded":false},{"timestamp":10.66,"box":[0.04635417088866234,0.16388888657093048,0.6364583931863308,0.43518517911434174],"occluded":false},{"timestamp":10.92,"box":[0.3057291805744171,0.14074073731899261,0.6375000178813934,0.41018517315387726],"occluded":false},{"timestamp":11.18,"box":[0.14375001192092896,0.11759258806705475,0.6427083611488342,0.388888880610466],"occluded":false},{"timestamp":11.44,"box":[0.2984375059604645,0.11296296119689941,0.6302083432674408,0.38055554032325745],"occluded":false},{"timestamp":11.700000000000001,"box":[0.31041666865348816,0.14629629254341125,0.6411458551883698,0.412962943315506],"occluded":false},{"timestamp":11.96,"box":[0.31041666865348816,0.3046296238899231,0.628125011920929,0.4277777671813965],"occluded":false},{"timestamp":12.22,"box":[0.4932291805744171,0.31388887763023376,0.6208333522081375,0.43888887763023376],"occluded":false},{"timestamp":12.48,"box":[0.49270835518836975,0.32314813137054443,0.6203125268220901,0.44907405972480774],"occluded":false},{"timestamp":12.74,"box":[0.5,0.32777777314186096,0.628125011920929,0.45185184478759766],"occluded":false},{"timestamp":13.0,"box":[0.5020833611488342,0.32777777314186096,0.6291667073965073,0.45092591643333435],"occluded":false},{"timestamp":13.26,"box":[0.49375003576278687,0.34351852536201477,0.6203125417232513,0.46851852536201477],"occluded":false},{"timestamp":13.52,"box":[0.49375003576278687,0.3759259283542633,0.6208333820104599,0.5018518567085266],"occluded":false},{"timestamp":13.780000000000001,"box":[0.5010417103767395,0.3768518567085266,0.6291667222976685,0.5037037134170532],"occluded":false},{"timestamp":14.040000000000001,"box":[0.503125011920929,0.3722222149372101,0.6307291835546494,0.4962962865829468],"occluded":false},{"timestamp":14.3,"box":[0.31041666865348816,0.37870368361473083,0.6317708492279053,0.5046296119689941],"occluded":false},{"timestamp":14.56,"box":[0.5052083730697632,0.37962961196899414,0.6317708790302277,0.5027777552604675],"occluded":false},{"timestamp":14.82,"box":[0.31041666865348816,0.364814817905426,0.6192708611488342,0.4824074059724808],"occluded":false},{"timestamp":15.08,"box":[0.39375001192092896,0.24074073135852814,0.6125000268220901,0.4722222089767456],"occluded":false},{"timestamp":15.34,"box":[0.5135416984558105,0.3444444239139557,0.6234375387430191,0.44999998062849045],"occluded":false},{"timestamp":15.6,"box":[0.5213541984558105,0.317592591047287,0.6234375387430191,0.4138888865709305],"occluded":false},{"timestamp":15.860000000000001,"box":[0.5151041746139526,0.3046296238899231,0.6078125163912773,0.39537036418914795],"occluded":false},{"timestamp":16.12,"box":[0.5093750357627869,0.3305555582046509,0.5927083715796471,0.40833333134651184],"occluded":false}],"mean_score":0.6831682920455933},{"noun_phrase":"large boulder","object_id":9,"presentation_id":3,"observations":[{"timestamp":2.08,"box":[0.4895833730697632,0.565740704536438,0.6098958775401115,0.7907406985759735],"occluded":false},{"timestamp":2.34,"box":[0.17135417461395264,0.40740740299224854,0.6338542103767395,0.8166666626930237],"occluded":false},{"timestamp":2.6,"box":[0.19895833730697632,0.42129629850387573,0.6645833551883698,0.7768518328666687],"occluded":false},{"timestamp":2.86,"box":[0.5489583611488342,0.5222222208976746,0.6781250387430191,0.7759259343147278],"occluded":false},{"timestamp":3.12,"box":[0.25468751788139343,0.44907405972480774,0.7250000536441803,0.8435184955596924],"occluded":false},{"timestamp":3.38,"box":[0.25468751788139343,0.46296295523643494,0.7281250357627869,0.8499999940395355],"occluded":false},{"timestamp":3.64,"box":[0.5531250238418579,0.6962962746620178,0.6937500238418579,0.9685184955596924],"occluded":false},{"timestamp":3.9,"box":[0.526562511920929,0.7157407402992249,0.673958346247673,0.970370352268219],"occluded":false},{"timestamp":4.16,"box":[0.004687500186264515,0.5564814805984497,0.7036458970978856,0.9694444239139557],"occluded":false},{"timestamp":4.42,"box":[0.5588542222976685,0.720370352268219,0.7145833969116211,0.970370352268219],"occluded":false},{"timestamp":4.68,"box":[0.5494791865348816,0.7555555701255798,0.7109375298023224,0.9703703820705414],"occluded":false},{"timestamp":4.94,"box":[0.5182291865348816,0.7361111044883728,0.677083358168602,0.970370352268219],"occluded":false},{"timestamp":5.2,"box":[0.15781250596046448,0.5324074029922485,0.6890625059604645,0.9694444239139557],"occluded":false},{"timestamp":5.46,"box":[0.53125,0.7509258985519409,0.7010416686534882,0.9703703373670578],"occluded":false},{"timestamp":5.72,"box":[0.534375011920929,0.7435185313224792,0.7026041895151138,0.9703703820705414],"occluded":false},{"timestamp":5.98,"box":[0.5250000357627869,0.7537037134170532,0.6984375417232513,0.9703703820705414],"occluded":false},{"timestamp":6.24,"box":[0.4921875298023224,0.7240740656852722,0.6687500327825546,0.9694444388151169],"occluded":false},{"timestamp":6.5,"box":[0.5281250476837158,0.6972222328186035,0.7031250596046448,0.9703703820705414],"occluded":false},{"timestamp":6.76,"box":[0.557812511920929,0.5148147940635681,0.7432291954755783,0.970370352268219],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.5625,0.690740704536438,0.7489583492279053,0.9712962508201599],"occluded":false},{"timestamp":7.28,"box":[0.550000011920929,0.800000011920929,0.7484375238418579,0.9712963104248047],"occluded":false},{"timestamp":7.54,"box":[0.43906253576278687,0.8064814805984497,0.741666704416275,0.9712962955236435],"occluded":false},{"timestamp":7.8,"box":[0.596875011920929,0.7259259223937988,0.8057291954755783,0.9703703671693802],"occluded":false},{"timestamp":8.06,"box":[0.24114584922790527,0.5601851940155029,0.8760417103767395,0.9703703820705414],"occluded":false},{"timestamp":8.32,"box":[0.19947917759418488,0.6175925731658936,0.9125000387430191,0.9712962806224823],"occluded":false},{"timestamp":8.58,"box":[0.6718750596046448,0.6425926089286804,0.9151042401790619,0.9712963104248047],"occluded":false},{"timestamp":8.84,"box":[0.39375001192092896,0.5324074029922485,0.9192708730697632,0.9712962806224823],"occluded":false},{"timestamp":9.1,"box":[0.4598958492279053,0.5564814805984497,0.9296875298023224,0.9712962806224823],"occluded":false},{"timestamp":9.36,"box":[0.4177083671092987,0.5287036895751953,0.991666704416275,0.9712962806224823],"occluded":false},{"timestamp":9.620000000000001,"box":[0.3895833492279053,0.5009258985519409,0.9994792342185974,0.9722221791744232],"occluded":false},{"timestamp":9.88,"box":[0.3369791805744171,0.4861111044883728,0.999479204416275,0.9712962806224823],"occluded":false},{"timestamp":10.14,"box":[0.3135416805744171,0.4166666567325592,0.999479204416275,0.9712962806224823],"occluded":false},{"timestamp":10.4,"box":[0.2927083373069763,0.4314814805984497,0.9994792342185974,0.9712963104248047],"occluded":false},{"timestamp":10.66,"box":[0.27916666865348816,0.3898147940635681,0.999479204416275,0.9712962508201599],"occluded":false},{"timestamp":10.92,"box":[0.27916666865348816,0.4037036895751953,0.999479204416275,0.9712962508201599],"occluded":false},{"timestamp":11.18,"box":[0.265625,0.4592592418193817,0.9994792342185974,0.9712962806224823],"occluded":false},{"timestamp":11.44,"box":[0.29218751192092896,0.3759259283542633,0.9994792342185974,0.9712962806224823],"occluded":false},{"timestamp":11.700000000000001,"box":[0.29322919249534607,0.345370352268219,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":11.96,"box":[0.27916666865348816,0.4037036895751953,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":12.22,"box":[0.2828125059604645,0.40740740299224854,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":12.48,"box":[0.29114586114883423,0.402777761220932,0.9994792342185974,0.9722221791744232],"occluded":false},{"timestamp":12.74,"box":[0.2645833492279053,0.4314814805984497,0.9994792342185974,0.9731481671333313],"occluded":false},{"timestamp":13.0,"box":[0.06041666865348816,0.28240740299224854,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":13.26,"box":[0.2645833492279053,0.42499998211860657,0.9994792342185974,0.9722221791744232],"occluded":false},{"timestamp":13.52,"box":[0.29218751192092896,0.4592592418193817,0.9994792342185974,0.9722221791744232],"occluded":false},{"timestamp":13.780000000000001,"box":[0.27916666865348816,0.47870370745658875,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":14.040000000000001,"box":[0.2786458432674408,0.4583333134651184,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":14.3,"box":[0.06041666865348816,0.36574074625968933,0.999479204416275,0.972222238779068],"occluded":false},{"timestamp":14.56,"box":[0.27916666865348816,0.4722222089767456,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":14.82,"box":[0.2786458432674408,0.49259257316589355,0.999479204416275,0.9712962806224823],"occluded":false},{"timestamp":15.08,"box":[0.24791668355464935,0.49259257316589355,0.9994792193174362,0.9712962806224823],"occluded":false},{"timestamp":15.34,"box":[0.31041666865348816,0.4407407343387604,0.999479204416275,0.9712962806224823],"occluded":false},{"timestamp":15.6,"box":[0.31666669249534607,0.3861111104488373,0.991666704416275,0.972222238779068],"occluded":false},{"timestamp":15.860000000000001,"box":[0.4489583671092987,0.3583333194255829,0.9494792520999908,0.9712962806224823],"occluded":false},{"timestamp":16.12,"box":[0.34427085518836975,0.3861111104488373,0.8838542401790619,0.9703703820705414],"occluded":false}],"mean_score":0.6831682920455933},{"noun_phrase":"large boulder","object_id":12,"presentation_id":4,"observations":[{"timestamp":10.14,"box":[0.0,0.720370352268219,0.5281250476837158,0.9712962806224823],"occluded":false},{"timestamp":10.4,"box":[0.0,0.75,0.5088542103767395,0.9712962955236435],"occluded":false},{"timestamp":10.66,"box":[0.0,0.7490740418434143,0.5171875357627869,0.9722221791744232],"occluded":false},{"timestamp":10.92,"box":[0.0,0.7314814925193787,0.5302083492279053,0.9722222238779068],"occluded":false},{"timestamp":11.18,"box":[0.0,0.7064814567565918,0.5250000357627869,0.9712962806224823],"occluded":false},{"timestamp":11.44,"box":[0.0,0.7027777433395386,0.5213541984558105,0.9712962508201599],"occluded":false},{"timestamp":11.700000000000001,"box":[0.0,0.6472222208976746,0.5328125357627869,0.9722222089767456],"occluded":false},{"timestamp":11.96,"box":[0.0,0.7453703284263611,0.5104166865348816,0.9722221791744232],"occluded":false},{"timestamp":12.22,"box":[0.0,0.7518518567085266,0.49635419249534607,0.9722222238779068],"occluded":false},{"timestamp":12.48,"box":[0.0,0.760185182094574,0.48906251788139343,0.9722222089767456],"occluded":false},{"timestamp":12.74,"box":[0.0,0.7620370388031006,0.4932291805744171,0.9722222238779068],"occluded":false},{"timestamp":13.0,"box":[0.0,0.7592592239379883,0.4906250238418579,0.9722221791744232],"occluded":false},{"timestamp":13.26,"box":[0.0,0.7759259343147278,0.468229204416275,0.9722222238779068],"occluded":false},{"timestamp":13.52,"box":[0.0,0.8064814805984497,0.4322916865348816,0.9722222238779068],"occluded":false},{"timestamp":13.780000000000001,"box":[0.0,0.7981481552124023,0.4364583492279053,0.9722222238779068],"occluded":false},{"timestamp":14.040000000000001,"box":[0.0,0.7907407283782959,0.43802085518836975,0.9722222089767456],"occluded":false},{"timestamp":14.3,"box":[0.0,0.7990740537643433,0.43958336114883423,0.9722221940755844],"occluded":false},{"timestamp":14.56,"box":[0.0,0.8009259104728699,0.44322919845581055,0.9722222089767456],"occluded":false},{"timestamp":14.82,"box":[0.0,0.79537034034729,0.4854166805744171,0.9722221940755844],"occluded":false},{"timestamp":15.08,"box":[0.0,0.7842592597007751,0.5140625238418579,0.9722222238779068],"occluded":false},{"timestamp":15.34,"box":[0.0,0.7222222089767456,0.53125,0.9712962806224823],"occluded":false},{"timestamp":15.6,"box":[0.0,0.5796296000480652,0.5416666865348816,0.9712962508201599],"occluded":false},{"timestamp":15.860000000000001,"box":[0.0,0.5601851940155029,0.5364583730697632,0.9712963104248047],"occluded":false},{"timestamp":16.12,"box":[0.0078125,0.5592592358589172,0.5296875238418579,0.9694444239139557],"occluded":false}],"mean_score":0.6604774594306946},{"noun_phrase":"large boulder","object_id":6,"presentation_id":5,"observations":[{"timestamp":1.3,"box":[0.5067708492279053,0.489814817905426,0.7229166924953461,0.9083333313465118],"occluded":false},{"timestamp":1.56,"box":[0.43385419249534607,0.47870370745658875,0.7072916924953461,0.9074074029922485],"occluded":false},{"timestamp":1.82,"box":[0.46666669845581055,0.5462962985038757,0.7645833790302277,0.970370352268219],"occluded":false},{"timestamp":2.08,"box":[0.3453125059604645,0.3583333194255829,0.8406250178813934,0.9694444239139557],"occluded":false},{"timestamp":2.34,"box":[0.5583333373069763,0.5361111164093018,0.8864583671092987,0.9712962806224823],"occluded":false},{"timestamp":2.6,"box":[0.19947917759418488,0.42129629850387573,0.9562500268220901,0.9666666388511658],"occluded":false},{"timestamp":2.86,"box":[0.6109375357627869,0.10833333432674408,0.9994792342185974,0.9990740865468979],"occluded":false},{"timestamp":3.12,"box":[0.6557291746139526,0.4833333194255829,0.999479204416275,0.9935185015201569],"occluded":false},{"timestamp":3.38,"box":[0.6473958492279053,0.12314814329147339,0.999479204416275,0.9925925731658936],"occluded":false},{"timestamp":3.64,"box":[0.3072916865348816,0.43518516421318054,0.9994792342185974,0.9990740716457367],"occluded":false},{"timestamp":3.9,"box":[0.24114584922790527,0.43518516421318054,0.9994792342185974,0.9990740716457367],"occluded":false},{"timestamp":4.16,"box":[0.3072916865348816,0.06666666269302368,0.9994792342185974,0.9990740418434143],"occluded":false},{"timestamp":4.42,"box":[0.6781250238418579,0.06666666269302368,0.999479204416275,0.9990740418434143],"occluded":false},{"timestamp":4.68,"box":[0.5151041746139526,0.3314814865589142,0.999479204416275,0.9990740716457367],"occluded":false},{"timestamp":4.94,"box":[0.6447917222976685,0.317592591047287,0.9994792342185974,0.9990740716457367],"occluded":false},{"timestamp":5.2,"box":[0.5015625357627869,0.317592591047287,0.9994792342185974,0.9990740716457367],"occluded":false},{"timestamp":5.46,"box":[0.29322919249534607,0.317592591047287,0.999479204416275,0.9907407462596893],"occluded":false},{"timestamp":5.72,"box":[0.6562500596046448,0.6370370388031006,0.9994792342185974,0.9759259223937988],"occluded":false},{"timestamp":5.98,"box":[0.6520833969116211,0.6481481194496155,0.9994792342185974,0.9990740418434143],"occluded":false},{"timestamp":6.24,"box":[0.5989583730697632,0.6212962865829468,0.9994792342185974,0.9740740656852722],"occluded":false},{"timestamp":6.5,"box":[0.6244791746139526,0.5694444179534912,0.999479204416275,0.9990740418434143],"occluded":false},{"timestamp":6.76,"box":[0.3072916865348816,0.4416666626930237,0.9994792342185974,0.9731481075286865],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.5291666984558105,0.5333333015441895,0.9994792342185974,0.9740740358829498],"occluded":false},{"timestamp":7.28,"box":[0.7500000596046448,0.6379629373550415,0.9994792342185974,0.9990740418434143],"occluded":false},{"timestamp":7.54,"box":[0.30677086114883423,0.5666666626930237,0.9994792342185974,0.9990740716457367],"occluded":false},{"timestamp":7.8,"box":[0.32083335518836975,0.5231481194496155,0.999479204416275,0.9722221791744232],"occluded":false},{"timestamp":8.06,"box":[0.45625001192092896,0.4546296298503876,0.9994792342185974,0.9990740716457367],"occluded":false},{"timestamp":8.32,"box":[0.612500011920929,0.33888888359069824,0.999479204416275,0.9990740418434143],"occluded":false},{"timestamp":8.58,"box":[0.9026042222976685,0.3592592477798462,0.9994792267680168,0.9731481075286865],"occluded":false},{"timestamp":8.84,"box":[0.22708334028720856,0.39629629254341125,0.9994791895151138,0.9759258925914764],"occluded":false},{"timestamp":9.1,"box":[0.18541668355464935,0.3481481373310089,0.9994792193174362,0.9712962806224823],"occluded":false},{"timestamp":9.36,"box":[0.98333340883255,0.4611110985279083,0.9994792435318232,0.7620370090007782],"occluded":false}],"mean_score":0.6559139490127563},{"noun_phrase":"large boulder","object_id":3,"presentation_id":6,"observations":[{"timestamp":0.0,"box":[0.04062500223517418,0.1537037044763565,0.20781251415610313,0.375],"occluded":false},{"timestamp":0.26,"box":[0.04218750074505806,0.21666666865348816,0.20364584401249886,0.4287036955356598],"occluded":false},{"timestamp":0.52,"box":[0.04843750223517418,0.17037037014961243,0.20625000819563866,0.3879629522562027],"occluded":false},{"timestamp":0.78,"box":[0.04635417088866234,0.138888880610466,0.2067708484828472,0.36574073135852814],"occluded":false},{"timestamp":1.04,"box":[0.03750000149011612,0.14166666567325592,0.19843750447034836,0.36759258806705475],"occluded":false},{"timestamp":1.3,"box":[0.0,0.07037036865949631,0.15052084624767303,0.32685183733701706],"occluded":false},{"timestamp":1.56,"box":[0.0,0.032407406717538834,0.12864583730697632,0.2898148037493229],"occluded":false},{"timestamp":1.82,"box":[0.0,0.1814814805984497,0.14739583432674408,0.42592592537403107],"occluded":false},{"timestamp":2.08,"box":[0.05937500298023224,0.20000000298023224,0.19739584624767303,0.4148148149251938],"occluded":false},{"timestamp":2.34,"box":[0.09062500298023224,0.19907407462596893,0.21979168057441711,0.42129629850387573],"occluded":false},{"timestamp":2.6,"box":[0.11145833879709244,0.21759258210659027,0.23541668057441711,0.4166666567325592],"occluded":false},{"timestamp":2.86,"box":[0.12135417014360428,0.22407406568527222,0.24322917312383652,0.4287036955356598],"occluded":false},{"timestamp":3.12,"box":[0.18333333730697632,0.25740739703178406,0.7828125357627869,0.5324074029922485],"occluded":false},{"timestamp":3.38,"box":[0.18020834028720856,0.27222222089767456,0.7692708522081375,0.5185185074806213],"occluded":false},{"timestamp":3.64,"box":[0.1354166716337204,0.35370370745658875,0.24895834177732468,0.5444444417953491],"occluded":false},{"timestamp":3.9,"box":[0.0911458358168602,0.36759257316589355,0.21458334475755692,0.5685184895992279],"occluded":false},{"timestamp":4.16,"box":[0.13125000894069672,0.3759259283542633,0.755208358168602,0.6712962985038757],"occluded":false},{"timestamp":4.42,"box":[0.13229167461395264,0.06759259104728699,0.99114590883255,0.6574073731899261],"occluded":false},{"timestamp":4.68,"box":[0.12552084028720856,0.06759259104728699,0.9911458641290665,0.6851851642131805],"occluded":false},{"timestamp":4.94,"box":[0.08281250298023224,0.3333333134651184,0.20468750596046448,0.5370370149612427],"occluded":false},{"timestamp":5.2,"box":[0.0911458358168602,0.32222220301628113,0.755208395421505,0.6574073731899261],"occluded":false},{"timestamp":5.46,"box":[0.07447917014360428,0.06759259104728699,0.9911458566784859,0.6574073731899261],"occluded":false},{"timestamp":5.72,"box":[0.10572917014360428,0.3314814865589142,0.7557292059063911,0.6435185074806213],"occluded":false},{"timestamp":5.98,"box":[0.11093750596046448,0.28703704476356506,0.22291667759418488,0.49166667461395264],"occluded":false},{"timestamp":6.24,"box":[0.0651041716337204,0.20370370149612427,0.18802084773778915,0.4296296238899231],"occluded":false},{"timestamp":6.5,"box":[0.07916667312383652,0.2666666507720947,0.5609375461935997,0.5185185074806213],"occluded":false},{"timestamp":6.76,"box":[0.11302083730697632,0.33796295523643494,0.22239584475755692,0.527777761220932],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.10468750447034836,0.3712962865829468,0.21510417759418488,0.5583333224058151],"occluded":false},{"timestamp":7.28,"box":[0.10312500596046448,0.42129629850387573,0.20937500894069672,0.6157407462596893],"occluded":false},{"timestamp":7.54,"box":[0.08489583432674408,0.402777761220932,0.19479167461395264,0.6009259074926376],"occluded":false},{"timestamp":7.8,"box":[0.13072916865348816,0.460185170173645,0.22968750447034836,0.6333333104848862],"occluded":false},{"timestamp":8.06,"box":[0.18020834028720856,0.5268518328666687,0.268229179084301,0.6935184895992279],"occluded":false},{"timestamp":8.32,"box":[0.1953125149011612,0.5249999761581421,0.28593751788139343,0.6620370149612427],"occluded":false},{"timestamp":8.58,"box":[0.19479167461395264,0.5074074268341064,0.2880208492279053,0.6620370596647263],"occluded":false},{"timestamp":8.84,"box":[0.18072918057441711,0.5,0.27135418355464935,0.6648148149251938],"occluded":false},{"timestamp":9.1,"box":[0.08645834028720856,0.432407408952713,0.20104167610406876,0.6240740716457367],"occluded":false},{"timestamp":9.36,"box":[0.0,0.3759259283542633,0.11927083879709244,0.6203703731298447],"occluded":false},{"timestamp":9.620000000000001,"box":[0.0,0.35185185074806213,0.05000000447034836,0.6361111104488373],"occluded":false},{"timestamp":15.860000000000001,"box":[0.0,0.21203702688217163,0.9963542222976685,0.32592591643333435],"occluded":false},{"timestamp":16.12,"box":[0.0,0.2916666567325592,0.0833333358168602,0.534259244799614],"occluded":false}],"mean_score":0.6952381134033203},{"noun_phrase":"large boulder","object_id":2,"presentation_id":7,"observations":[{"timestamp":0.0,"box":[0.2343750149011612,0.21574074029922485,0.3192708492279053,0.33888888359069824],"occluded":false},{"timestamp":0.26,"box":[0.23489584028720856,0.2629629671573639,0.3203125149011612,0.3787037059664726],"occluded":false},{"timestamp":0.52,"box":[0.24062500894069672,0.2351851761341095,0.3265625163912773,0.35462962090969086],"occluded":false},{"timestamp":0.78,"box":[0.24270834028720856,0.20277777314186096,0.3291666805744171,0.32592591643333435],"occluded":false},{"timestamp":1.04,"box":[0.14375001192092896,0.21203702688217163,0.3239583522081375,0.36574073135852814],"occluded":false},{"timestamp":1.3,"box":[0.20104168355464935,0.1814814805984497,0.2916666865348816,0.3157407343387604],"occluded":false},{"timestamp":1.56,"box":[0.18020834028720856,0.14722222089767456,0.27395834773778915,0.2888888865709305],"occluded":false},{"timestamp":1.82,"box":[0.20937500894069672,0.27407407760620117,0.2968750149011612,0.4018518477678299],"occluded":false},{"timestamp":2.08,"box":[0.15781250596046448,0.26759257912635803,0.33802084624767303,0.42129628360271454],"occluded":false},{"timestamp":2.34,"box":[0.18541668355464935,0.27314814925193787,0.35833336412906647,0.42129629850387573],"occluded":false},{"timestamp":2.6,"box":[0.2942708432674408,0.2638888955116272,0.37552084773778915,0.38055556267499924],"occluded":false},{"timestamp":2.86,"box":[0.29322919249534607,0.060185182839632034,0.38333336263895035,0.3749999888241291],"occluded":false},{"timestamp":3.12,"box":[0.3489583432674408,0.3018518388271332,0.4270833507180214,0.41018517315387726],"occluded":false},{"timestamp":3.38,"box":[0.27916666865348816,0.31296294927597046,0.42500001192092896,0.42129628360271454],"occluded":false},{"timestamp":3.64,"box":[0.3114583492279053,0.41018518805503845,0.3906250223517418,0.5240740776062012],"occluded":false},{"timestamp":3.9,"box":[0.20937500894069672,0.4305555522441864,0.36250001192092896,0.5472222194075584],"occluded":false},{"timestamp":4.16,"box":[0.31041666865348816,0.4407407343387604,0.38854167610406876,0.5555555447936058],"occluded":false},{"timestamp":4.42,"box":[0.3140625059604645,0.4342592656612396,0.6854166984558105,0.7268518507480621],"occluded":false},{"timestamp":4.68,"box":[0.23750001192092896,0.43796294927597046,0.38750001788139343,0.5638888776302338],"occluded":false},{"timestamp":4.94,"box":[0.2802083492279053,0.4203703701496124,0.35989585518836975,0.5398148149251938],"occluded":false},{"timestamp":5.2,"box":[0.28697916865348816,0.40925925970077515,0.36770834028720856,0.5277777761220932],"occluded":false},{"timestamp":5.46,"box":[0.29322919249534607,0.4342592656612396,0.3744791969656944,0.555555559694767],"occluded":false},{"timestamp":5.72,"box":[0.29895836114883423,0.4203703701496124,0.3796875327825546,0.5407407358288765],"occluded":false},{"timestamp":5.98,"box":[0.30416667461395264,0.39814814925193787,0.3812500089406967,0.5222222208976746],"occluded":false},{"timestamp":6.24,"box":[0.27395835518836975,0.3472222089767456,0.35364586114883423,0.47499997913837433],"occluded":false},{"timestamp":6.5,"box":[0.28958335518836975,0.36944442987442017,0.3697916939854622,0.48981479555368423],"occluded":false},{"timestamp":6.76,"box":[0.3125000298023224,0.4148148000240326,0.39479169994592667,0.5305555388331413],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.20937500894069672,0.4175925850868225,0.3916666805744171,0.5472222119569778],"occluded":false},{"timestamp":7.28,"box":[0.30416667461395264,0.5027777552604675,0.385416679084301,0.6222222000360489],"occluded":false},{"timestamp":7.54,"box":[0.29114586114883423,0.4972222149372101,0.37395836412906647,0.6175925806164742],"occluded":false},{"timestamp":7.8,"box":[0.32760417461395264,0.5074074268341064,0.4114583432674408,0.6166666820645332],"occluded":false},{"timestamp":8.06,"box":[0.008333333767950535,0.5425925850868225,0.45052085164934397,0.645370364189148],"occluded":false},{"timestamp":8.32,"box":[0.38333335518836975,0.5185185074806213,0.46822918951511383,0.6138888746500015],"occluded":false},{"timestamp":8.58,"box":[0.3838541805744171,0.5138888955116272,0.4677083492279053,0.6120370402932167],"occluded":false},{"timestamp":8.84,"box":[0.008333333767950535,0.5046296119689941,0.4567708456888795,0.6231481283903122],"occluded":false},{"timestamp":9.1,"box":[0.18541668355464935,0.44907405972480774,0.42083336412906647,0.5777777582406998],"occluded":false},{"timestamp":9.36,"box":[0.03281250223517418,0.4120370149612427,0.38593750819563866,0.5546296089887619],"occluded":false},{"timestamp":9.620000000000001,"box":[0.20989584922790527,0.3861111104488373,0.35312502086162567,0.5555555522441864],"occluded":false},{"timestamp":9.88,"box":[0.1640625149011612,0.36759257316589355,0.325520858168602,0.5601851642131805],"occluded":false},{"timestamp":10.14,"box":[0.11354167014360428,0.29999998211860657,0.6020833626389503,0.5120370090007782],"occluded":false},{"timestamp":10.4,"box":[0.06822916865348816,0.2925925850868225,0.25468751788139343,0.6370370090007782],"occluded":false},{"timestamp":10.66,"box":[0.008333333767950535,0.2648148238658905,0.25520835164934397,0.6157407462596893],"occluded":false},{"timestamp":10.92,"box":[0.05260416865348816,0.24537037312984467,0.2552083432674408,0.581481471657753],"occluded":false},{"timestamp":11.18,"box":[0.0,0.2231481373310089,0.25729167461395264,0.6046296060085297],"occluded":false},{"timestamp":11.44,"box":[0.04322917014360428,0.21851851046085358,0.504687525331974,0.460185170173645],"occluded":false},{"timestamp":11.700000000000001,"box":[0.05781250447034836,0.2537037134170532,0.2562500163912773,0.49074074625968933],"occluded":false},{"timestamp":11.96,"box":[0.04270833730697632,0.25925925374031067,0.2421875149011612,0.4999999850988388],"occluded":false},{"timestamp":12.22,"box":[0.03489583358168602,0.2666666507720947,0.23541667684912682,0.5092592388391495],"occluded":false},{"timestamp":12.48,"box":[0.03437500074505806,0.27407407760620117,0.23489584401249886,0.5175925940275192],"occluded":false},{"timestamp":12.74,"box":[0.04375000298023224,0.2787036895751953,0.24322918057441711,0.5203703492879868],"occluded":false},{"timestamp":13.0,"box":[0.0468750037252903,0.2759259343147278,0.9770833887159824,0.7407407462596893],"occluded":false},{"timestamp":13.26,"box":[0.0364583358168602,0.19907407462596893,0.45625003427267075,0.5324073880910873],"occluded":false},{"timestamp":13.52,"box":[0.03750000149011612,0.22685185074806213,0.48906252533197403,0.5629629492759705],"occluded":false},{"timestamp":13.780000000000001,"box":[0.04791666939854622,0.22685185074806213,0.40833334997296333,0.5583333373069763],"occluded":false},{"timestamp":14.040000000000001,"box":[0.04895833507180214,0.22685185074806213,0.42916667833924294,0.5509259104728699],"occluded":false},{"timestamp":14.3,"box":[0.05052083730697632,0.22685185074806213,0.421354204416275,0.5601851642131805],"occluded":false},{"timestamp":14.56,"box":[0.05312500149011612,0.24074073135852814,0.39375001937150955,0.5620370358228683],"occluded":false},{"timestamp":14.82,"box":[0.06093750149011612,0.24074073135852814,0.3802083507180214,0.5574073940515518],"occluded":false},{"timestamp":15.08,"box":[0.07760416716337204,0.23981481790542603,0.56041669100523,0.5555555522441864],"occluded":false},{"timestamp":15.34,"box":[0.1328125,0.22685185074806213,0.5744791924953461,0.5194444358348846],"occluded":false},{"timestamp":15.6,"box":[0.1744791716337204,0.3083333373069763,0.49895836412906647,0.4861111044883728],"occluded":false},{"timestamp":15.860000000000001,"box":[0.2031250149011612,0.22685185074806213,0.4697916954755783,0.4703703671693802],"occluded":false},{"timestamp":16.12,"box":[0.23125001788139343,0.25462961196899414,0.5604166984558105,0.47499997913837433],"occluded":false}],"mean_score":0.7655677795410156},{"noun_phrase":"large boulder","object_id":13,"presentation_id":8,"observations":[{"timestamp":11.44,"box":[0.2979166805744171,0.1101851835846901,0.5270833522081375,0.3768518343567848],"occluded":false},{"timestamp":11.700000000000001,"box":[0.31041666865348816,0.14537036418914795,0.6020833551883698,0.40925925970077515],"occluded":false},{"timestamp":11.96,"box":[0.2958333492279053,0.15555554628372192,0.616666704416275,0.4296296238899231],"occluded":false},{"timestamp":12.22,"box":[0.04635417088866234,0.1648148149251938,0.6026042066514492,0.43796296417713165],"occluded":false},{"timestamp":12.48,"box":[0.2880208492279053,0.17314814031124115,0.5177083611488342,0.4407407194375992],"occluded":false},{"timestamp":12.74,"box":[0.29635417461395264,0.1787036955356598,0.5260416865348816,0.45277777314186096],"occluded":false},{"timestamp":13.0,"box":[0.06041666865348816,0.17777776718139648,0.5260416865348816,0.45185184478759766],"occluded":false},{"timestamp":13.26,"box":[0.10208334028720856,0.192592591047287,0.520833358168602,0.46759259700775146],"occluded":false},{"timestamp":13.52,"box":[0.28958335518836975,0.22592592239379883,0.5203125327825546,0.4990740716457367],"occluded":false},{"timestamp":13.780000000000001,"box":[0.06041666865348816,0.22499999403953552,0.5260416865348816,0.4990740716457367],"occluded":false},{"timestamp":14.040000000000001,"box":[0.12968750298023224,0.21851851046085358,0.5744792073965073,0.49166665971279144],"occluded":false},{"timestamp":14.3,"box":[0.06041666865348816,0.22592592239379883,0.6020833551883698,0.4990740716457367],"occluded":false},{"timestamp":14.56,"box":[0.07447917014360428,0.22962962090969086,0.6161458566784859,0.5018518418073654],"occluded":false},{"timestamp":14.82,"box":[0.07395833730697632,0.22685185074806213,0.6026042103767395,0.48055556416511536],"occluded":false},{"timestamp":15.08,"box":[0.15781250596046448,0.23333333432674408,0.5750000178813934,0.4712962955236435],"occluded":false},{"timestamp":15.34,"box":[0.1432291716337204,0.22407406568527222,0.5885417014360428,0.4425925761461258],"occluded":false},{"timestamp":15.6,"box":[0.18541668355464935,0.21203702688217163,0.6161458641290665,0.40925924479961395],"occluded":false},{"timestamp":15.860000000000001,"box":[0.3750000298023224,0.21296295523643494,0.5885417014360428,0.3916666507720947],"occluded":false},{"timestamp":16.12,"box":[0.3843750059604645,0.2490740716457367,0.5328125208616257,0.4009259194135666],"occluded":false}],"mean_score":0.7730495929718018},{"noun_phrase":"large boulder","object_id":5,"presentation_id":9,"observations":[{"timestamp":0.26,"box":[0.00364583358168602,0.3166666626930237,0.16302084550261497,0.45370370149612427],"occluded":false},{"timestamp":0.52,"box":[0.00729166716337204,0.2648148238658905,0.16302084177732468,0.4148148149251938],"occluded":false},{"timestamp":0.78,"box":[0.00729166716337204,0.2351851761341095,0.16718750447034836,0.3861110955476761],"occluded":false},{"timestamp":1.04,"box":[0.0,0.23981481790542603,0.15937501192092896,0.3898148089647293],"occluded":false},{"timestamp":1.3,"box":[0.0,0.18703703582286835,0.10729166865348816,0.35185185074806213],"occluded":false},{"timestamp":1.56,"box":[0.0,0.14999999105930328,0.12656250596046448,0.591666653752327],"occluded":false},{"timestamp":1.82,"box":[0.0,0.28425925970077515,0.11510417610406876,0.446296289563179],"occluded":false},{"timestamp":2.08,"box":[0.006770833861082792,0.2888888716697693,0.1692708428017795,0.4425925761461258],"occluded":false},{"timestamp":2.34,"box":[0.03593750298023224,0.28148147463798523,0.19062501192092896,0.43611110746860504],"occluded":false},{"timestamp":2.6,"box":[0.06302084028720856,0.26851850748062134,0.21666668355464935,0.4462962746620178],"occluded":false},{"timestamp":2.86,"box":[0.04531250149011612,0.21944443881511688,0.22447917610406876,0.44999998807907104],"occluded":false},{"timestamp":3.12,"box":[0.13906250894069672,0.32777777314186096,0.27031251788139343,0.4740740656852722],"occluded":false},{"timestamp":3.38,"box":[0.13333334028720856,0.3499999940395355,0.26875001192092896,0.4879629611968994],"occluded":false},{"timestamp":3.64,"box":[0.0598958358168602,0.4314814805984497,0.2317708507180214,0.7611111104488373],"occluded":false},{"timestamp":3.9,"box":[0.0,0.4037036895751953,0.2135416716337204,0.8101851642131805],"occluded":false},{"timestamp":4.16,"box":[0.0,0.42129629850387573,0.21041667461395264,0.7620370388031006],"occluded":false},{"timestamp":4.42,"box":[0.0,0.44907405972480774,0.2135416716337204,0.7824073731899261],"occluded":false},{"timestamp":4.68,"box":[0.05572916939854622,0.45185184478759766,0.20208333805203438,0.6055555492639542],"occluded":false},{"timestamp":4.94,"box":[0.0,0.39907407760620117,0.1614583432674408,0.5787037014961243],"occluded":false},{"timestamp":5.2,"box":[0.0,0.36574074625968933,0.17083334922790527,0.567592591047287],"occluded":false},{"timestamp":5.46,"box":[0.01770833507180214,0.4148148000240326,0.17500001564621925,0.595370352268219],"occluded":false},{"timestamp":5.72,"box":[0.02395833469927311,0.4000000059604645,0.17968750931322575,0.5694444477558136],"occluded":false},{"timestamp":5.98,"box":[0.004687500186264515,0.3166666626930237,0.17343750316649675,0.5249999910593033],"occluded":false},{"timestamp":6.24,"box":[0.0,0.24074073135852814,0.5604166984558105,0.7407407313585281],"occluded":false},{"timestamp":6.5,"box":[0.0,0.28333333134651184,0.1588541716337204,0.7129629552364349],"occluded":false},{"timestamp":6.76,"box":[0.03437500074505806,0.3759259283542633,0.18854166939854622,0.5814814865589142],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.02500000223517418,0.412962943315506,0.18385417386889458,0.6101851612329483],"occluded":false},{"timestamp":7.28,"box":[0.0,0.45092591643333435,0.16718751192092896,0.6611111015081406],"occluded":false},{"timestamp":7.54,"box":[0.0,0.4342592656612396,0.15156251192092896,0.6518518477678299],"occluded":false},{"timestamp":7.8,"box":[0.02239583432674408,0.5046296119689941,0.20156250894069672,0.6972222030162811],"occluded":false},{"timestamp":8.06,"box":[0.05000000447034836,0.5824074149131775,0.25208335369825363,0.7555555552244186],"occluded":false},{"timestamp":8.32,"box":[0.10572917014360428,0.5907407402992249,0.6854167059063911,0.7472222149372101],"occluded":false},{"timestamp":8.58,"box":[0.09843750298023224,0.5703703761100769,0.2718750089406967,0.7240740805864334],"occluded":false},{"timestamp":8.84,"box":[0.0703125,0.5666666626930237,0.25312501192092896,0.7277777642011642],"occluded":false},{"timestamp":9.1,"box":[0.0,0.4990740716457367,0.1796875149011612,0.7037037014961243],"occluded":false},{"timestamp":9.36,"box":[0.0,0.520370364189148,0.09010417014360428,0.7083333283662796],"occluded":false}],"mean_score":0.675126850605011},{"noun_phrase":"large boulder","object_id":1,"presentation_id":10,"observations":[{"timestamp":0.0,"box":[0.6536458730697632,0.032407406717538834,0.9994792342185974,0.9666666425764561],"occluded":false},{"timestamp":0.26,"box":[0.6484375596046448,0.031481482088565826,0.9994792342185974,0.9333333149552345],"occluded":false},{"timestamp":0.52,"box":[0.6760417222976685,0.03055555559694767,0.9994792342185974,0.9574073646217585],"occluded":false},{"timestamp":0.78,"box":[0.06041666865348816,0.03055555559694767,0.999479204416275,0.9037036988884211],"occluded":false},{"timestamp":1.04,"box":[0.14375001192092896,0.03055555559694767,0.9994792342185974,0.8759259078651667],"occluded":false},{"timestamp":1.3,"box":[0.11614584177732468,0.031481482088565826,0.9994792267680168,0.9277777448296547],"occluded":false},{"timestamp":1.56,"box":[0.22708334028720856,0.03055555559694767,0.9994791895151138,0.9629629347473383],"occluded":false},{"timestamp":1.82,"box":[0.5880208611488342,0.03055555559694767,0.999479204416275,0.8925925586372614],"occluded":false},{"timestamp":2.08,"box":[0.26875001192092896,0.03055555559694767,0.9994792342185974,0.914814779534936],"occluded":false},{"timestamp":2.34,"box":[0.29635417461395264,0.03055555559694767,0.9994792342185974,0.5462963078171015],"occluded":false},{"timestamp":2.6,"box":[0.9458333849906921,0.031481482088565826,0.9994792193174362,0.18703702837228775],"occluded":false},{"timestamp":3.9,"box":[0.9895833730697632,0.031481482088565826,0.9994792072102427,0.06296296417713165],"occluded":false},{"timestamp":15.860000000000001,"box":[0.004687500186264515,0.28240740299224854,0.9994792463257909,0.970370352268219],"occluded":false},{"timestamp":16.12,"box":[0.07343750447034836,0.19907407462596893,0.9994792267680168,0.9722221940755844],"occluded":false}],"mean_score":0.9179486632347107},{"noun_phrase":"large boulder","object_id":7,"presentation_id":11,"observations":[{"timestamp":1.82,"box":[0.28958335518836975,0.35462963581085205,0.5057291984558105,0.6314814686775208],"occluded":false},{"timestamp":2.08,"box":[0.14375001192092896,0.33888888359069824,0.6994792222976685,0.9148147702217102],"occluded":false},{"timestamp":2.34,"box":[0.17135417461395264,0.3481481373310089,0.5057291984558105,0.6249999701976776],"occluded":false},{"timestamp":2.6,"box":[0.19895833730697632,0.32592591643333435,0.5296875238418579,0.5916666388511658],"occluded":false},{"timestamp":2.86,"box":[0.21302084624767303,0.10925925523042679,0.9911458939313889,0.5879629626870155],"occluded":false},{"timestamp":3.12,"box":[0.24114584922790527,0.3629629611968994,0.7546875476837158,0.6370370388031006],"occluded":false},{"timestamp":3.38,"box":[0.25468751788139343,0.1787036955356598,0.9911458790302277,0.6444444358348846],"occluded":false},{"timestamp":3.64,"box":[0.32447919249534607,0.4749999940395355,0.5458333641290665,0.7583333253860474],"occluded":false},{"timestamp":3.9,"box":[0.14375001192092896,0.4962962865829468,0.5744791924953461,0.7962962687015533],"occluded":false},{"timestamp":4.16,"box":[0.004687500186264515,0.3314814865589142,0.9911458613350987,0.8101851940155029],"occluded":false},{"timestamp":4.42,"box":[0.38593751192092896,0.4981481432914734,0.6994791924953461,0.7962962985038757],"occluded":false},{"timestamp":4.68,"box":[0.07395833730697632,0.3314814865589142,0.99114590883255,0.8379629552364349],"occluded":false},{"timestamp":4.94,"box":[0.06041666865348816,0.317592591047287,0.9911458790302277,0.8101851642131805],"occluded":false},{"timestamp":5.2,"box":[0.01875000074505806,0.40740740299224854,0.8661458976566792,0.9212962985038757],"occluded":false},{"timestamp":5.46,"box":[0.03281250223517418,0.43518516421318054,0.589062537997961,0.9490740597248077],"occluded":false},{"timestamp":5.72,"box":[0.16770833730697632,0.49351850152015686,0.6437500417232513,0.8101851642131805],"occluded":false},{"timestamp":5.98,"box":[0.03229166939854622,0.37962961196899414,0.6020833440124989,0.8101851642131805],"occluded":false},{"timestamp":6.24,"box":[0.03281250223517418,0.35185185074806213,0.5473958514630795,0.7685185074806213],"occluded":false},{"timestamp":6.5,"box":[0.08802083879709244,0.3731481432914734,0.8942708745598793,0.8657407164573669],"occluded":false},{"timestamp":6.76,"box":[0.08802083879709244,0.43518516421318054,0.617187537252903,0.8796296119689941],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.38333335518836975,0.49166664481163025,0.7552083730697632,0.8101851642131805],"occluded":false},{"timestamp":7.28,"box":[0.15729168057441711,0.5712962746620178,0.616666704416275,0.9212962687015533],"occluded":false},{"timestamp":7.54,"box":[0.12916667759418488,0.5462962985038757,0.7687500268220901,0.8935185074806213],"occluded":false},{"timestamp":7.8,"box":[0.4046875238418579,0.5518518686294556,0.6447917073965073,0.9351851940155029],"occluded":false},{"timestamp":8.06,"box":[0.15781250596046448,0.5648148059844971,0.6994791924953461,0.8574073910713196],"occluded":false},{"timestamp":8.32,"box":[0.18541668355464935,0.36574074625968933,0.9354167431592941,0.8027777671813965],"occluded":false},{"timestamp":8.58,"box":[0.1848958432674408,0.39351850748062134,0.9354167282581329,0.8166666626930237],"occluded":false},{"timestamp":8.84,"box":[0.17135417461395264,0.5472221970558167,0.7276042103767395,0.8472221791744232],"occluded":false},{"timestamp":9.1,"box":[0.15781250596046448,0.49259257316589355,0.7411458790302277,0.8231481313705444],"occluded":false},{"timestamp":9.36,"box":[0.37968751788139343,0.45185184478759766,0.991666704416275,0.9351851642131805],"occluded":false},{"timestamp":9.620000000000001,"box":[0.10572917014360428,0.10185185074806213,0.996354229748249,0.9074073731899261],"occluded":false},{"timestamp":9.88,"box":[0.14375001192092896,0.09537036716938019,0.99114590883255,0.9500000029802322],"occluded":false},{"timestamp":10.14,"box":[0.11250000447034836,0.09537036716938019,0.9973958656191826,0.9527777582406998],"occluded":false},{"timestamp":10.4,"box":[0.03593750298023224,0.317592591047287,0.9973958879709244,0.9537036716938019],"occluded":false},{"timestamp":10.66,"box":[0.02916666865348816,0.33796295523643494,0.9963541924953461,0.9638888537883759],"occluded":false},{"timestamp":10.92,"box":[0.08802083879709244,0.345370352268219,0.9958333745598793,0.9462962746620178],"occluded":false},{"timestamp":11.18,"box":[0.01458333432674408,0.32314813137054443,0.6343750208616257,0.9194444417953491],"occluded":false},{"timestamp":11.44,"box":[0.0,0.3157407343387604,0.9494792222976685,0.9111110866069794],"occluded":false},{"timestamp":11.700000000000001,"box":[0.0,0.3490740656852722,0.9770833849906921,0.9472222328186035],"occluded":false},{"timestamp":11.96,"box":[0.03281250223517418,0.3611111044883728,0.9635417126119137,0.9611110687255859],"occluded":false},{"timestamp":12.22,"box":[0.02916666865348816,0.33796295523643494,0.9848959147930145,0.9694444239139557],"occluded":false},{"timestamp":12.48,"box":[0.2161458432674408,0.37962961196899414,0.6119791865348816,0.9694443941116333],"occluded":false},{"timestamp":12.74,"box":[0.03281250223517418,0.35185185074806213,0.9921875260770321,0.9712962806224823],"occluded":false},{"timestamp":13.0,"box":[0.03281250223517418,0.38333332538604736,0.9770834110677242,0.970370352268219],"occluded":false},{"timestamp":13.26,"box":[0.07083334028720856,0.39722222089767456,0.9562500268220901,0.9712963104248047],"occluded":false},{"timestamp":13.52,"box":[0.03229166939854622,0.39351850748062134,0.9552083797752857,0.9712962508201599],"occluded":false},{"timestamp":13.780000000000001,"box":[0.04218750074505806,0.4277777671813965,0.9911458976566792,0.970370352268219],"occluded":false},{"timestamp":14.040000000000001,"box":[0.10208334028720856,0.42129629850387573,0.9916667491197586,0.9712963104248047],"occluded":false},{"timestamp":14.3,"box":[0.0,0.4296296238899231,0.9776042103767395,0.970370352268219],"occluded":false},{"timestamp":14.56,"box":[0.0,0.36574074625968933,0.9505208730697632,0.9703703820705414],"occluded":false},{"timestamp":14.82,"box":[0.0,0.4009259045124054,0.9885417222976685,0.9712962806224823],"occluded":false},{"timestamp":15.08,"box":[0.0,0.4000000059604645,0.9953125715255737,0.9666666686534882],"occluded":false},{"timestamp":15.34,"box":[0.11614584177732468,0.37962961196899414,0.8255208656191826,0.9037036895751953],"occluded":false},{"timestamp":15.6,"box":[0.0,0.2481481432914734,0.9916667342185974,0.9564814567565918],"occluded":false},{"timestamp":15.860000000000001,"box":[0.2223958522081375,0.345370352268219,0.9911458641290665,0.8870370388031006],"occluded":false},{"timestamp":16.12,"box":[0.26875001192092896,0.3768518567085266,0.8520833849906921,0.7416666746139526],"occluded":false}],"mean_score":0.6908212304115295},{"noun_phrase":"large boulder","object_id":0,"presentation_id":12,"observations":[{"timestamp":0.0,"box":[0.17343750596046448,0.24722221493721008,0.35729168355464935,0.4824073910713196],"occluded":false},{"timestamp":0.26,"box":[0.17291668057441711,0.2944444417953491,0.36093752086162567,0.52407406270504],"occluded":false},{"timestamp":0.52,"box":[0.17604167759418488,0.260185182094574,0.3630208522081375,0.5018518418073654],"occluded":false},{"timestamp":0.78,"box":[0.04635417088866234,0.22870369255542755,0.3645833395421505,0.470370352268219],"occluded":false},{"timestamp":1.04,"box":[0.06041666865348816,0.23240740597248077,0.35572919249534607,0.47870369255542755],"occluded":false},{"timestamp":1.3,"box":[0.12031250447034836,0.19074073433876038,0.3255208507180214,0.4749999940395355],"occluded":false},{"timestamp":1.56,"box":[0.06041666865348816,0.14444443583488464,0.6302083432674408,0.8796295821666718],"occluded":false},{"timestamp":1.82,"box":[0.12708334624767303,0.2796296179294586,0.3307291865348816,0.5583333075046539],"occluded":false},{"timestamp":2.08,"box":[0.1302083432674408,0.2768518328666687,0.7270833551883698,0.9148147702217102],"occluded":false},{"timestamp":2.34,"box":[0.10208334028720856,0.2768518328666687,0.8104167133569717,0.6990740597248077],"occluded":false},{"timestamp":2.6,"box":[0.19895833730697632,0.27314814925193787,0.4140625149011612,0.5259259343147278],"occluded":false},{"timestamp":2.86,"box":[0.21302084624767303,0.012037036940455437,0.9911458939313889,0.5462962966412306],"occluded":false},{"timestamp":3.12,"box":[0.25468751788139343,0.3055555522441864,0.7552084028720856,0.5611110925674438],"occluded":false},{"timestamp":3.38,"box":[0.25468751788139343,0.10925925523042679,0.9911458790302277,0.5703703537583351],"occluded":false},{"timestamp":3.64,"box":[0.23125001788139343,0.345370352268219,0.9911458790302277,0.6787036657333374],"occluded":false},{"timestamp":3.9,"box":[0.004687500186264515,0.35185185074806213,0.39479167480021715,0.7018518447875977],"occluded":false},{"timestamp":4.16,"box":[0.2223958522081375,0.06759259104728699,0.9911458641290665,0.7111110985279083],"occluded":false},{"timestamp":4.42,"box":[0.22343751788139343,0.06759259104728699,0.9911458790302277,0.7018518149852753],"occluded":false},{"timestamp":4.68,"box":[0.06041666865348816,0.06759259104728699,0.9911458790302277,0.8379629552364349],"occluded":false},{"timestamp":4.94,"box":[0.004687500186264515,0.06759259104728699,0.9911458613350987,0.6888888776302338],"occluded":false},{"timestamp":5.2,"box":[0.0182291679084301,0.06759259104728699,0.9911458902060986,0.6749999821186066],"occluded":false},{"timestamp":5.46,"box":[0.03281250223517418,0.06759259104728699,0.9911458753049374,0.7018518149852753],"occluded":false},{"timestamp":5.72,"box":[0.0468750037252903,0.31018519401550293,0.40260419622063637,0.6814814805984497],"occluded":false},{"timestamp":5.98,"box":[0.03281250223517418,0.26851850748062134,0.3984375260770321,0.662962943315506],"occluded":false},{"timestamp":6.24,"box":[0.11614584177732468,0.28240740299224854,0.3682292029261589,0.6231481432914734],"occluded":false},{"timestamp":6.5,"box":[0.1692708432674408,0.3203703761100769,0.7552084028720856,0.6333333253860474],"occluded":false},{"timestamp":6.76,"box":[0.0885416716337204,0.33796295523643494,0.41718752682209015,0.6675925850868225],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.18958334624767303,0.39629629254341125,0.41718752682209015,0.6861110925674438],"occluded":false},{"timestamp":7.28,"box":[0.17812500894069672,0.45277777314186096,0.7828125208616257,0.767592579126358],"occluded":false},{"timestamp":7.54,"box":[0.01875000074505806,0.40740740299224854,0.8661458976566792,0.7685185074806213],"occluded":false},{"timestamp":7.8,"box":[0.20364584028720856,0.4731481373310089,0.4364583492279053,0.7555555403232574],"occluded":false},{"timestamp":8.06,"box":[0.15781250596046448,0.5231481194496155,0.6270833611488342,0.7833333015441895],"occluded":false},{"timestamp":8.32,"box":[0.1432291716337204,0.36574074625968933,0.9354166835546494,0.7462962865829468],"occluded":false},{"timestamp":8.58,"box":[0.14375001192092896,0.39351850748062134,0.9354166984558105,0.7435185015201569],"occluded":false},{"timestamp":8.84,"box":[0.12916667759418488,0.49537035822868347,0.49010418355464935,0.7611110806465149],"occluded":false},{"timestamp":9.1,"box":[0.01875000074505806,0.4277777671813965,0.45625003054738045,0.7333333194255829],"occluded":false},{"timestamp":9.36,"box":[0.03229166939854622,0.3722222149372101,0.42916669324040413,0.7416666448116302],"occluded":false},{"timestamp":9.620000000000001,"box":[0.0,0.3324074149131775,0.5989583730697632,0.7731481492519379],"occluded":false},{"timestamp":9.88,"box":[0.0,0.3018518388271332,0.6411458849906921,0.7972221970558167],"occluded":false},{"timestamp":10.14,"box":[0.0,0.23981481790542603,0.5744792222976685,0.7592592239379883],"occluded":false},{"timestamp":10.4,"box":[0.0,0.2796296179294586,0.99114590883255,0.7805555164813995],"occluded":false},{"timestamp":10.66,"box":[0.0,0.2648148238658905,0.99114590883255,0.7759259045124054],"occluded":false},{"timestamp":10.92,"box":[0.0,0.25,0.612500011920929,0.7611110806465149],"occluded":false},{"timestamp":11.18,"box":[0.0,0.22499999403953552,0.5848958492279053,0.737962931394577],"occluded":false},{"timestamp":11.44,"box":[0.0,0.22685185074806213,0.5854166746139526,0.7333333194255829],"occluded":false},{"timestamp":11.700000000000001,"box":[0.0,0.25648146867752075,0.5848958492279053,0.7638888955116272],"occluded":false},{"timestamp":11.96,"box":[0.0,0.26851850748062134,0.5994791984558105,0.7740740776062012],"occluded":false},{"timestamp":12.22,"box":[0.0,0.26851850748062134,0.5572916865348816,0.7842592597007751],"occluded":false},{"timestamp":12.48,"box":[0.0,0.3009259104728699,0.28697916865348816,0.7916666567325592],"occluded":false},{"timestamp":12.74,"box":[0.0,0.28240740299224854,0.5989583730697632,0.7972221970558167],"occluded":false},{"timestamp":13.0,"box":[0.0,0.28240740299224854,0.2958333492279053,0.7944444417953491],"occluded":false},{"timestamp":13.26,"box":[0.0,0.29629629850387573,0.28541669249534607,0.8101851940155029],"occluded":false},{"timestamp":13.52,"box":[0.0,0.33703702688217163,0.557812511920929,0.8425925970077515],"occluded":false},{"timestamp":13.780000000000001,"box":[0.0,0.3157407343387604,0.5057291984558105,0.8361110985279083],"occluded":false},{"timestamp":14.040000000000001,"box":[0.0,0.3055555522441864,0.47812503576278687,0.8305555284023285],"occluded":false},{"timestamp":14.3,"box":[0.0,0.31203702092170715,0.557812511920929,0.8398148119449615],"occluded":false},{"timestamp":14.56,"box":[0.0,0.3194444477558136,0.5744792222976685,0.837037056684494],"occluded":false},{"timestamp":14.82,"box":[0.0,0.32777777314186096,0.5848958492279053,0.893518477678299],"occluded":false},{"timestamp":15.08,"box":[0.0,0.32314813137054443,0.5989583730697632,0.9148147702217102],"occluded":false},{"timestamp":15.34,"box":[0.0,0.24444444477558136,0.6130208373069763,0.7962963134050369],"occluded":false},{"timestamp":15.6,"box":[0.004166666883975267,0.24074073135852814,0.6130208414979279,0.8092592507600784],"occluded":false},{"timestamp":15.860000000000001,"box":[0.0078125,0.25925925374031067,0.5989583730697632,0.6564814746379852],"occluded":false},{"timestamp":16.12,"box":[0.0442708358168602,0.2898148000240326,0.5437500402331352,0.6453703343868256],"occluded":false}],"mean_score":0.7852349281311035}],"dropped":{"over_max_tracklets":6},"fates":[{"noun_phrase":"large boulder","object_id":0,"frames":63,"mean_score":0.7852349281311035,"mean_area":0.2585558647021176,"outcome":"kept","presentation_id":12,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"large boulder","object_id":1,"frames":14,"mean_score":0.9179486632347107,"mean_area":0.4833720999137583,"outcome":"kept","presentation_id":10,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"large boulder","object_id":2,"frames":63,"mean_score":0.7655677795410156,"mean_area":0.05263657115736005,"outcome":"kept","presentation_id":7,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"large boulder","object_id":3,"frames":40,"mean_score":0.6952381134033203,"mean_area":0.08738174101698809,"outcome":"kept","presentation_id":6,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"large boulder","object_id":4,"frames":63,"mean_score":0.6831682920455933,"mean_area":0.062948252274052,"outcome":"kept","presentation_id":2,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"large boulder","object_id":5,"frames":36,"mean_score":0.675126850605011,"mean_area":0.0431319807587681,"outcome":"kept","presentation_id":9,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"large boulder","object_id":6,"frames":32,"mean_score":0.6559139490127563,"mean_area":0.2669789807329256,"outcome":"kept","presentation_id":5,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"large boulder","object_id":7,"frames":56,"mean_score":0.6908212304115295,"mean_area":0.3771535929633307,"outcome":"kept","presentation_id":11,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"large boulder","object_id":8,"frames":56,"mean_score":0.8048780560493469,"mean_area":0.23556023065215928,"outcome":"kept","presentation_id":1,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"large boulder","object_id":9,"frames":55,"mean_score":0.6831682920455933,"mean_area":0.24452945404979004,"outcome":"kept","presentation_id":3,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"large boulder","object_id":10,"frames":23,"mean_score":0.7299578189849854,"mean_area":0.004173586130247751,"outcome":"over_max_tracklets","presentation_id":null,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"large boulder","object_id":12,"frames":24,"mean_score":0.6604774594306946,"mean_area":0.12125038654660107,"outcome":"kept","presentation_id":4,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"large boulder","object_id":13,"frames":19,"mean_score":0.7730495929718018,"mean_area":0.09626482787006188,"outcome":"kept","presentation_id":8,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"white sign","object_id":0,"frames":13,"mean_score":0.8552036285400391,"mean_area":0.0009850576056952638,"outcome":"over_max_tracklets","presentation_id":null,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"white sign","object_id":1,"frames":28,"mean_score":0.8632478713989258,"mean_area":0.021021981362315617,"outcome":"over_max_tracklets","presentation_id":null,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"white sign","object_id":2,"frames":61,"mean_score":0.941605806350708,"mean_area":0.024296330345504245,"outcome":"over_max_tracklets","presentation_id":null,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"white sign","object_id":3,"frames":28,"mean_score":0.8202247023582458,"mean_area":0.0006681616716298336,"outcome":"over_max_tracklets","presentation_id":null,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"white sign","object_id":4,"frames":30,"mean_score":0.8632478713989258,"mean_area":0.0009179044881173967,"outcome":"over_max_tracklets","presentation_id":null,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null}],"gates":{"min_track_frames":3,"min_mean_score":0.5,"merge_iou":0.8,"max_tracklets":12}}},"render":{"inputs":{"tracklets":12,"as_video":true,"fps":1.0,"max_frames":30},"outputs":{"media":"/home/zzk/data/video_bbox/vlm_gemini3.5_lite/renders/--cif7Nv6BI_chunk_006_scene_028_choices.mp4","media_bytes":4508091,"frames":[{"index":1,"timestamp":0.0,"file":"--cif7Nv6BI_chunk_006_scene_028_choices.mp4"},{"index":2,"timestamp":1.04,"file":"--cif7Nv6BI_chunk_006_scene_028_choices.mp4"},{"index":3,"timestamp":2.08,"file":"--cif7Nv6BI_chunk_006_scene_028_choices.mp4"},{"index":4,"timestamp":3.12,"file":"--cif7Nv6BI_chunk_006_scene_028_choices.mp4"},{"index":5,"timestamp":4.16,"file":"--cif7Nv6BI_chunk_006_scene_028_choices.mp4"},{"index":6,"timestamp":4.94,"file":"--cif7Nv6BI_chunk_006_scene_028_choices.mp4"},{"index":7,"timestamp":5.98,"file":"--cif7Nv6BI_chunk_006_scene_028_choices.mp4"},{"index":8,"timestamp":7.0200000000000005,"file":"--cif7Nv6BI_chunk_006_scene_028_choices.mp4"},{"index":9,"timestamp":8.06,"file":"--cif7Nv6BI_chunk_006_scene_028_choices.mp4"},{"index":10,"timestamp":9.1,"file":"--cif7Nv6BI_chunk_006_scene_028_choices.mp4"},{"index":11,"timestamp":10.14,"file":"--cif7Nv6BI_chunk_006_scene_028_choices.mp4"},{"index":12,"timestamp":11.18,"file":"--cif7Nv6BI_chunk_006_scene_028_choices.mp4"},{"index":13,"timestamp":11.96,"file":"--cif7Nv6BI_chunk_006_scene_028_choices.mp4"},{"index":14,"timestamp":13.0,"file":"--cif7Nv6BI_chunk_006_scene_028_choices.mp4"},{"index":15,"timestamp":14.040000000000001,"file":"--cif7Nv6BI_chunk_006_scene_028_choices.mp4"},{"index":16,"timestamp":15.08,"file":"--cif7Nv6BI_chunk_006_scene_028_choices.mp4"},{"index":17,"timestamp":16.12,"file":"--cif7Nv6BI_chunk_006_scene_028_choices.mp4"}]}},"select":{"inputs":{"model":"gemini-3.5-flash-lite","temperature":0.0,"as_video":true,"media_bytes":4508091,"media":["--cif7Nv6BI_chunk_006_scene_028_choices.mp4"],"choice_frames":17,"system_prompt":"You are labelling video data for referring multi-object tracking.\n\nYou will see a numbered sequence of frames from one clip, in time order. Each frame has its own number and timestamp printed in the top-left corner. Candidate objects have been detected for you and drawn on the frames as coloured boxes, each labelled with a number. The same object keeps the same number and the same colour on every frame. Below each frame is a strip of enlarged crops of those same numbered boxes, so you can identify a small or distant object.\n\nFor each description you are given, answer two questions:\n\n1. MATCHING_IDS — which of the numbered objects the description refers to. List every number that matches, or an EMPTY LIST if none of them do. An empty list is a correct and expected answer: some descriptions are deliberately about something that is not in this clip. Do not pick a number just to have picked one, and do not pick a number for an object that merely resembles the description.\n\n2. START_FRAME and END_FRAME — the frame numbers between which the described state actually holds. If the description is \"the car that's turning left\", the interval is the frames in which that car is turning, NOT every frame in which the car is visible. Use the frame numbers printed on the frames. START_FRAME must not be greater than END_FRAME. When MATCHING_IDS is empty, give the full range (the first and last frame numbers); it is ignored.\n\n3. CONFIDENCE — how sure you are of the MATCHING_IDS answer, from 0.0 to 1.0. Be honest: a low number costs the example nothing, and an overconfident wrong answer becomes a wrong label.\n\nRules:\n- Only use numbers that are actually drawn on the frames.\n- Never output coordinates, box values, or pixel positions.\n- Judge motion by comparing the object's position across frames, not by guessing from one frame.\n","user_prompt":"The video has 17 frames, numbered 1 to 17. Each frame is one moment of the clip and has its number burned into the top left corner.\n\nAnswer for each of the following descriptions, in the order given:\n1. the white information sign posted near the stone wall\n2. the large boulders piled up against the blue sky","questions":["the white information sign posted near the stone wall","the large boulders piled up against the blue sky"]},"outputs":{"raw_response":"{\n  \"answers\": [\n    {\n      \"description_index\": 1,\n      \"matching_ids\": [],\n      \"start_frame\": 1,\n      \"end_frame\": 17,\n      \"confidence\": 0.0\n    },\n    {\n      \"description_index\": 2,\n      \"matching_ids\": [2],\n      \"start_frame\": 1,\n      \"end_frame\": 17,\n      \"confidence\": 0.95\n    }\n  ]\n}","raw_answers":[{"description_index":1,"matching_ids":[],"start_frame":1,"end_frame":17,"confidence":0.0},{"description_index":2,"matching_ids":[2],"start_frame":1,"end_frame":17,"confidence":0.95}],"answers":[{"expression":"the white information sign posted near the stone wall","matching_presentation_ids":[],"start_choice_frame":1,"end_choice_frame":17,"confidence":0.0,"self_reported_confidence":0.0},{"expression":"the large boulders piled up against the blue sky","matching_presentation_ids":[2],"start_choice_frame":1,"end_choice_frame":17,"confidence":0.95,"self_reported_confidence":0.95}]}},"verify":{"inputs":{"expressions":2,"tracklets":12,"selections":2,"gates":"interval, count, unknown ids, confidence, then geometry (motion)"},"outputs":{"kept":[],"rejected":{"the white information sign posted near the stone wall":"count_mismatch","the large boulders piled up against the blue sky":"count_mismatch"},"rejection_counts":{"count_mismatch":2},"row":null}}}},{"index":7,"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_014_scene_013.mp4","status":"complete","steps":{"decode":{"inputs":{"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_014_scene_013.mp4","sample_fps":4.0},"outputs":{"width":1920,"height":1080,"source_fps":50.0,"duration_seconds":17.86,"frames_kept":69,"blank_frames_dropped":0,"timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88,10.14,10.4,10.66,10.92,11.18,11.44,11.700000000000001,11.96,12.22,12.48,12.74,13.0,13.26,13.52,13.780000000000001,14.040000000000001,14.3,14.56,14.82,15.08,15.34,15.6,15.860000000000001,16.12,16.38,16.64,16.9,17.16,17.42,17.68],"error":""}},"propose":{"inputs":{"model":"gemini-3.5-flash-lite","temperature":0.4,"video_bytes":21892411,"system_prompt":"You are annotating video clips to build training data for referring multi-object tracking: given a video and a natural-language description, a model must output the boxes of every object matching that description, over time.\n\nYour job is to write the descriptions. You do NOT output coordinates, boxes, pixel positions, or any numbers about where things are. A separate detector finds the objects.\n\nFor every clip you produce a list of items. Each item has two texts that must NOT be the same:\n\nEXPRESSION — what a person would say to single out the object or objects. It may describe motion, temporal order, interaction with other objects, and spatial relations. It should read like something a person would actually say.\n  Good: \"the black car that's turning left\"\n  Good: \"the cyclist who overtakes the bus\"\n  Good: \"the two people who walk out of frame on the right\"\n\nNOUN_PHRASE — the same object as a plain, timeless visual category with its visible attributes, and NOTHING else. The detector is prompted with this text and looks at ONE frame at a time, so anything about time, order, or action is useless to it and makes it fail.\n  Allowed: category nouns, colours, materials, clothing, size, type\n  Forbidden: verbs and participles of action (\"turning\", \"walks\", \"is moving\"), temporal words (\"then\", \"starts\", \"after\", \"first\", \"last\"), relative clauses (\"that ...\", \"which ...\", \"who ...\"), ordinals and rank words (\"first\", \"leftmost\", \"nearest\"), and anything over six words.\n  \"the black car that's turning left\" -> \"black car\"\n  \"the cyclist who overtakes the bus\" -> \"cyclist\"\n  \"the two people who walk out of frame on the right\" -> \"person\"\n\nKIND — what makes the expression hard: \"motion\", \"appearance\", \"spatial\", or \"interaction\". Aim for about 35% \"motion\" items: motion is the part a single-frame model cannot already solve, so it is the most valuable, but a list of only motion items is not usable training data either.\n\nEXPECTED_COUNT — how many distinct objects in the clip match the EXPRESSION. Count carefully; this is checked against the detector's answer, and a wrong count discards the item.\n\nIS_NEGATIVE — at most ONE item may be a negative, and only when you also write at least one non-negative item. A negative is an expression that recombines attributes that ARE present in the clip into a whole that is NOT. If the only black car is moving, \"the parked black car\" is a good negative. Its NOUN_PHRASE must still be the category that IS present (\"black car\"), because that is what makes the absence checkable rather than merely unverified. A negative's EXPECTED_COUNT is 0.\n\nRules that discard an item, so follow them exactly:\n- Never write coordinates, box values, pixel positions, or percentages of the frame.\n- Never let a NOUN_PHRASE contain a verb of action, a temporal word, a relative clause, or an ordinal.\n- Never write an EXPRESSION whose object is not visible in the clip, unless it is marked IS_NEGATIVE.\n- Prefer objects that are clearly visible and identifiable over background clutter.\n- Do not describe the camera or the shot; describe objects in the scene.\n","user_prompt":"Watch this clip and write between 1 and 3 items following the rules. Cover different objects rather than restating one object several ways, and include any negative in the same list.\n"},"outputs":{"raw_response":"{\n  \"items\": [\n    {\n      \"expression\": \"the woman talking in front of the shop\",\n      \"noun_phrase\": \"woman\",\n      \"kind\": \"appearance\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the hand waving across the camera view\",\n      \"noun_phrase\": \"hand\",\n      \"kind\": \"motion\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    }\n  ]\n}","raw_items":[{"expression":"the woman talking in front of the shop","noun_phrase":"woman","kind":"appearance","expected_count":1,"is_negative":false},{"expression":"the hand waving across the camera view","noun_phrase":"hand","kind":"motion","expected_count":1,"is_negative":false}],"accepted":[{"expression":"the woman talking in front of the shop","noun_phrase":"woman","kind":"appearance","expected_count":1,"is_negative":false},{"expression":"the hand waving across the camera view","noun_phrase":"hand","kind":"motion","expected_count":1,"is_negative":false}],"gated":[]}},"track":{"inputs":{"tracker":"Sam31MultiplexTracker","model_id":"/mnt/ceph3/ec/jacky/sam3.1","noun_phrases":["hand","woman"],"expressions_in":2,"frames":69,"sample_fps":4.0,"frame_size":"1920x1080","timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88,10.14,10.4,10.66,10.92,11.18,11.44,11.700000000000001,11.96,12.22,12.48,12.74,13.0,13.26,13.52,13.780000000000001,14.040000000000001,14.3,14.56,14.82,15.08,15.34,15.6,15.860000000000001,16.12,16.38,16.64,16.9,17.16,17.42,17.68],"session":"one session over 69 JPEG frames; then per phrase: reset_session, add_prompt(text) on frame 0, propagate_in_video","passes":2},"outputs":{"raw_tracks":[{"noun_phrase":"hand","object_id":0,"observations":[{"timestamp":1.04,"box":[0.05677083507180214,0.7268518209457397,0.22500001266598701,0.9712962657213211],"occluded":false},{"timestamp":1.3,"box":[0.04114583507180214,0.5009258985519409,0.37031251564621925,0.9712962508201599],"occluded":false},{"timestamp":1.56,"box":[0.0937500074505806,0.45370370149612427,0.7411458566784859,0.9712963104248047],"occluded":false},{"timestamp":1.82,"box":[0.09947916865348816,0.4611110985279083,0.5604166984558105,0.9722221791744232],"occluded":false},{"timestamp":2.08,"box":[0.06093750149011612,0.3898147940635681,0.7583333626389503,0.9712962508201599],"occluded":false},{"timestamp":2.34,"box":[0.0598958358168602,0.402777761220932,0.4135416969656944,0.9712962806224823],"occluded":false},{"timestamp":2.6,"box":[0.04843750223517418,0.4148148000240326,0.7031250260770321,0.9712962806224823],"occluded":false}],"scores":[0.8827838897705078,0.8827838897705078,0.8827838897705078,0.8827838897705078,0.8827838897705078,0.8827838897705078,0.8827838897705078]},{"noun_phrase":"woman","object_id":0,"observations":[{"timestamp":0.0,"box":[0.2994791865348816,0.17129629850387573,0.7786458730697632,0.9712963104248047],"occluded":false},{"timestamp":0.26,"box":[0.30937501788139343,0.1787036955356598,0.7463541924953461,0.9722221791744232],"occluded":false},{"timestamp":0.52,"box":[0.2942708432674408,0.182407408952713,0.7125000357627869,0.9731481373310089],"occluded":false},{"timestamp":0.78,"box":[0.3135416805744171,0.19814814627170563,0.7161458730697632,0.9740740805864334],"occluded":false},{"timestamp":1.04,"box":[0.11562500894069672,0.23148147761821747,0.7416667193174362,0.974074050784111],"occluded":false},{"timestamp":1.3,"box":[0.37916669249534607,0.260185182094574,0.7546875476837158,0.9731481075286865],"occluded":false},{"timestamp":1.56,"box":[0.19947917759418488,0.287962943315506,0.7682292014360428,0.9731481373310089],"occluded":false},{"timestamp":1.82,"box":[0.22708334028720856,0.2657407224178314,0.7718750387430191,0.9722221791744232],"occluded":false},{"timestamp":2.08,"box":[0.40781253576278687,0.2657407224178314,0.7843750417232513,0.9722221791744232],"occluded":false},{"timestamp":2.34,"box":[0.4140625298023224,0.2481481432914734,0.8000000417232513,0.9722222089767456],"occluded":false},{"timestamp":2.6,"box":[0.17135417461395264,0.22962962090969086,0.8088542222976685,0.9712962955236435],"occluded":false},{"timestamp":2.86,"box":[0.41822919249534607,0.21944443881511688,0.8083333671092987,0.9722221940755844],"occluded":false},{"timestamp":3.12,"box":[0.3916666805744171,0.20370370149612427,0.7895833551883698,0.9722222089767456],"occluded":false},{"timestamp":3.38,"box":[0.37291669845581055,0.18703703582286835,0.7848958969116211,0.9722221940755844],"occluded":false},{"timestamp":3.64,"box":[0.37135419249534607,0.16018518805503845,0.7901042103767395,0.9712962806224823],"occluded":false},{"timestamp":3.9,"box":[0.36250001192092896,0.15648147463798523,0.7791666984558105,0.9731481373310089],"occluded":false},{"timestamp":4.16,"box":[0.3265625238418579,0.17222221195697784,0.7406250536441803,0.9740740209817886],"occluded":false},{"timestamp":4.42,"box":[0.31197917461395264,0.1935185194015503,0.7145833671092987,0.9740740656852722],"occluded":false},{"timestamp":4.68,"box":[0.2984375059604645,0.19907407462596893,0.6927083730697632,0.9731481522321701],"occluded":false},{"timestamp":4.94,"box":[0.2848958373069763,0.18518517911434174,0.679166704416275,0.9731481522321701],"occluded":false},{"timestamp":5.2,"box":[0.28177085518836975,0.1814814805984497,0.6770833730697632,0.9740740656852722],"occluded":false},{"timestamp":5.46,"box":[0.2760416865348816,0.18703703582286835,0.6755208671092987,0.9750000089406967],"occluded":false},{"timestamp":5.72,"box":[0.2578125,0.19074073433876038,0.6744791865348816,0.9740740358829498],"occluded":false},{"timestamp":5.98,"box":[0.26979169249534607,0.19629628956317902,0.6817708909511566,0.9759258776903152],"occluded":false},{"timestamp":6.24,"box":[0.2828125059604645,0.192592591047287,0.694791704416275,0.9759258925914764],"occluded":false},{"timestamp":6.5,"box":[0.2968750298023224,0.17685185372829437,0.7161458730697632,0.9722221940755844],"occluded":false},{"timestamp":6.76,"box":[0.30156251788139343,0.1657407432794571,0.7281250357627869,0.9731481224298477],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.29479166865348816,0.14999999105930328,0.7270833551883698,0.9722222238779068],"occluded":false},{"timestamp":7.28,"box":[0.2942708432674408,0.14351852238178253,0.7364583611488342,0.9722222238779068],"occluded":false},{"timestamp":7.54,"box":[0.30156251788139343,0.13518518209457397,0.7473958730697632,0.9731481075286865],"occluded":false},{"timestamp":7.8,"box":[0.3083333373069763,0.14999999105930328,0.7562500238418579,0.9731481224298477],"occluded":false},{"timestamp":8.06,"box":[0.30781251192092896,0.1509259194135666,0.7520833611488342,0.9722221940755844],"occluded":false},{"timestamp":8.32,"box":[0.30781251192092896,0.1509259194135666,0.7468750476837158,0.9731481522321701],"occluded":false},{"timestamp":8.58,"box":[0.29635417461395264,0.13611111044883728,0.7437500357627869,0.9722221791744232],"occluded":false},{"timestamp":8.84,"box":[0.2942708432674408,0.138888880610466,0.7385416924953461,0.9722221940755844],"occluded":false},{"timestamp":9.1,"box":[0.29114586114883423,0.14351852238178253,0.7312500476837158,0.9731481224298477],"occluded":false},{"timestamp":9.36,"box":[0.2786458432674408,0.14537036418914795,0.7151041924953461,0.9731481075286865],"occluded":false},{"timestamp":9.620000000000001,"box":[0.2734375,0.15555554628372192,0.7026041746139526,0.9722222089767456],"occluded":false},{"timestamp":9.88,"box":[0.2614583373069763,0.1546296328306198,0.6895833611488342,0.9731481522321701],"occluded":false},{"timestamp":10.14,"box":[0.2526041865348816,0.1537037044763565,0.6786458790302277,0.9731481224298477],"occluded":false},{"timestamp":10.4,"box":[0.25052085518836975,0.14537036418914795,0.6812500357627869,0.9722222089767456],"occluded":false},{"timestamp":10.66,"box":[0.24687501788139343,0.16111110150814056,0.6760416924953461,0.9777777642011642],"occluded":false},{"timestamp":10.92,"box":[0.2473958432674408,0.18333333730697632,0.6802083551883698,0.9740740656852722],"occluded":false},{"timestamp":11.18,"box":[0.24635417759418488,0.16296295821666718,0.6833333522081375,0.974074050784111],"occluded":false},{"timestamp":11.44,"box":[0.2343750149011612,0.14814814925193787,0.680208370089531,0.9749999940395355],"occluded":false},{"timestamp":11.700000000000001,"box":[0.26875001192092896,0.16388888657093048,0.7161458730697632,0.9731481224298477],"occluded":false},{"timestamp":11.96,"box":[0.2671875059604645,0.1787036955356598,0.7140625417232513,0.9740740358829498],"occluded":false},{"timestamp":12.22,"box":[0.2723958492279053,0.17685185372829437,0.7161458730697632,0.9731481522321701],"occluded":false},{"timestamp":12.48,"box":[0.2848958373069763,0.182407408952713,0.7265625298023224,0.9740740358829498],"occluded":false},{"timestamp":12.74,"box":[0.3187500238418579,0.18981480598449707,0.7656250596046448,0.9740740656852722],"occluded":false},{"timestamp":13.0,"box":[0.3609375059604645,0.1527777761220932,0.8343750238418579,0.9712962955236435],"occluded":false},{"timestamp":13.26,"box":[0.41979169845581055,0.13333332538604736,0.8947917222976685,0.9722222089767456],"occluded":false},{"timestamp":13.52,"box":[0.4208333492279053,0.1379629671573639,0.8656250536441803,0.9731481373310089],"occluded":false},{"timestamp":13.780000000000001,"box":[0.35572919249534607,0.17592592537403107,0.7588542103767395,0.9722222238779068],"occluded":false},{"timestamp":14.040000000000001,"box":[0.2526041865348816,0.182407408952713,0.6328125298023224,0.972222238779068],"occluded":false},{"timestamp":14.3,"box":[0.1302083432674408,0.1527777761220932,0.5536458790302277,0.9722221940755844],"occluded":false},{"timestamp":14.56,"box":[0.09270834177732468,0.10462962836027145,0.5416667088866234,0.972222201526165],"occluded":false},{"timestamp":14.82,"box":[0.0,0.028703702613711357,0.4427083432674408,0.971296263858676],"occluded":false},{"timestamp":15.08,"box":[0.0,0.028703702613711357,0.13385418057441711,0.9749999772757292],"occluded":false}],"scores":[0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917]}],"tracklets":[{"noun_phrase":"hand","object_id":0,"presentation_id":1,"observations":[{"timestamp":1.04,"box":[0.05677083507180214,0.7268518209457397,0.22500001266598701,0.9712962657213211],"occluded":false},{"timestamp":1.3,"box":[0.04114583507180214,0.5009258985519409,0.37031251564621925,0.9712962508201599],"occluded":false},{"timestamp":1.56,"box":[0.0937500074505806,0.45370370149612427,0.7411458566784859,0.9712963104248047],"occluded":false},{"timestamp":1.82,"box":[0.09947916865348816,0.4611110985279083,0.5604166984558105,0.9722221791744232],"occluded":false},{"timestamp":2.08,"box":[0.06093750149011612,0.3898147940635681,0.7583333626389503,0.9712962508201599],"occluded":false},{"timestamp":2.34,"box":[0.0598958358168602,0.402777761220932,0.4135416969656944,0.9712962806224823],"occluded":false},{"timestamp":2.6,"box":[0.04843750223517418,0.4148148000240326,0.7031250260770321,0.9712962806224823],"occluded":false}],"mean_score":0.8827838897705078},{"noun_phrase":"woman","object_id":0,"presentation_id":2,"observations":[{"timestamp":0.0,"box":[0.2994791865348816,0.17129629850387573,0.7786458730697632,0.9712963104248047],"occluded":false},{"timestamp":0.26,"box":[0.30937501788139343,0.1787036955356598,0.7463541924953461,0.9722221791744232],"occluded":false},{"timestamp":0.52,"box":[0.2942708432674408,0.182407408952713,0.7125000357627869,0.9731481373310089],"occluded":false},{"timestamp":0.78,"box":[0.3135416805744171,0.19814814627170563,0.7161458730697632,0.9740740805864334],"occluded":false},{"timestamp":1.04,"box":[0.11562500894069672,0.23148147761821747,0.7416667193174362,0.974074050784111],"occluded":false},{"timestamp":1.3,"box":[0.37916669249534607,0.260185182094574,0.7546875476837158,0.9731481075286865],"occluded":false},{"timestamp":1.56,"box":[0.19947917759418488,0.287962943315506,0.7682292014360428,0.9731481373310089],"occluded":false},{"timestamp":1.82,"box":[0.22708334028720856,0.2657407224178314,0.7718750387430191,0.9722221791744232],"occluded":false},{"timestamp":2.08,"box":[0.40781253576278687,0.2657407224178314,0.7843750417232513,0.9722221791744232],"occluded":false},{"timestamp":2.34,"box":[0.4140625298023224,0.2481481432914734,0.8000000417232513,0.9722222089767456],"occluded":false},{"timestamp":2.6,"box":[0.17135417461395264,0.22962962090969086,0.8088542222976685,0.9712962955236435],"occluded":false},{"timestamp":2.86,"box":[0.41822919249534607,0.21944443881511688,0.8083333671092987,0.9722221940755844],"occluded":false},{"timestamp":3.12,"box":[0.3916666805744171,0.20370370149612427,0.7895833551883698,0.9722222089767456],"occluded":false},{"timestamp":3.38,"box":[0.37291669845581055,0.18703703582286835,0.7848958969116211,0.9722221940755844],"occluded":false},{"timestamp":3.64,"box":[0.37135419249534607,0.16018518805503845,0.7901042103767395,0.9712962806224823],"occluded":false},{"timestamp":3.9,"box":[0.36250001192092896,0.15648147463798523,0.7791666984558105,0.9731481373310089],"occluded":false},{"timestamp":4.16,"box":[0.3265625238418579,0.17222221195697784,0.7406250536441803,0.9740740209817886],"occluded":false},{"timestamp":4.42,"box":[0.31197917461395264,0.1935185194015503,0.7145833671092987,0.9740740656852722],"occluded":false},{"timestamp":4.68,"box":[0.2984375059604645,0.19907407462596893,0.6927083730697632,0.9731481522321701],"occluded":false},{"timestamp":4.94,"box":[0.2848958373069763,0.18518517911434174,0.679166704416275,0.9731481522321701],"occluded":false},{"timestamp":5.2,"box":[0.28177085518836975,0.1814814805984497,0.6770833730697632,0.9740740656852722],"occluded":false},{"timestamp":5.46,"box":[0.2760416865348816,0.18703703582286835,0.6755208671092987,0.9750000089406967],"occluded":false},{"timestamp":5.72,"box":[0.2578125,0.19074073433876038,0.6744791865348816,0.9740740358829498],"occluded":false},{"timestamp":5.98,"box":[0.26979169249534607,0.19629628956317902,0.6817708909511566,0.9759258776903152],"occluded":false},{"timestamp":6.24,"box":[0.2828125059604645,0.192592591047287,0.694791704416275,0.9759258925914764],"occluded":false},{"timestamp":6.5,"box":[0.2968750298023224,0.17685185372829437,0.7161458730697632,0.9722221940755844],"occluded":false},{"timestamp":6.76,"box":[0.30156251788139343,0.1657407432794571,0.7281250357627869,0.9731481224298477],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.29479166865348816,0.14999999105930328,0.7270833551883698,0.9722222238779068],"occluded":false},{"timestamp":7.28,"box":[0.2942708432674408,0.14351852238178253,0.7364583611488342,0.9722222238779068],"occluded":false},{"timestamp":7.54,"box":[0.30156251788139343,0.13518518209457397,0.7473958730697632,0.9731481075286865],"occluded":false},{"timestamp":7.8,"box":[0.3083333373069763,0.14999999105930328,0.7562500238418579,0.9731481224298477],"occluded":false},{"timestamp":8.06,"box":[0.30781251192092896,0.1509259194135666,0.7520833611488342,0.9722221940755844],"occluded":false},{"timestamp":8.32,"box":[0.30781251192092896,0.1509259194135666,0.7468750476837158,0.9731481522321701],"occluded":false},{"timestamp":8.58,"box":[0.29635417461395264,0.13611111044883728,0.7437500357627869,0.9722221791744232],"occluded":false},{"timestamp":8.84,"box":[0.2942708432674408,0.138888880610466,0.7385416924953461,0.9722221940755844],"occluded":false},{"timestamp":9.1,"box":[0.29114586114883423,0.14351852238178253,0.7312500476837158,0.9731481224298477],"occluded":false},{"timestamp":9.36,"box":[0.2786458432674408,0.14537036418914795,0.7151041924953461,0.9731481075286865],"occluded":false},{"timestamp":9.620000000000001,"box":[0.2734375,0.15555554628372192,0.7026041746139526,0.9722222089767456],"occluded":false},{"timestamp":9.88,"box":[0.2614583373069763,0.1546296328306198,0.6895833611488342,0.9731481522321701],"occluded":false},{"timestamp":10.14,"box":[0.2526041865348816,0.1537037044763565,0.6786458790302277,0.9731481224298477],"occluded":false},{"timestamp":10.4,"box":[0.25052085518836975,0.14537036418914795,0.6812500357627869,0.9722222089767456],"occluded":false},{"timestamp":10.66,"box":[0.24687501788139343,0.16111110150814056,0.6760416924953461,0.9777777642011642],"occluded":false},{"timestamp":10.92,"box":[0.2473958432674408,0.18333333730697632,0.6802083551883698,0.9740740656852722],"occluded":false},{"timestamp":11.18,"box":[0.24635417759418488,0.16296295821666718,0.6833333522081375,0.974074050784111],"occluded":false},{"timestamp":11.44,"box":[0.2343750149011612,0.14814814925193787,0.680208370089531,0.9749999940395355],"occluded":false},{"timestamp":11.700000000000001,"box":[0.26875001192092896,0.16388888657093048,0.7161458730697632,0.9731481224298477],"occluded":false},{"timestamp":11.96,"box":[0.2671875059604645,0.1787036955356598,0.7140625417232513,0.9740740358829498],"occluded":false},{"timestamp":12.22,"box":[0.2723958492279053,0.17685185372829437,0.7161458730697632,0.9731481522321701],"occluded":false},{"timestamp":12.48,"box":[0.2848958373069763,0.182407408952713,0.7265625298023224,0.9740740358829498],"occluded":false},{"timestamp":12.74,"box":[0.3187500238418579,0.18981480598449707,0.7656250596046448,0.9740740656852722],"occluded":false},{"timestamp":13.0,"box":[0.3609375059604645,0.1527777761220932,0.8343750238418579,0.9712962955236435],"occluded":false},{"timestamp":13.26,"box":[0.41979169845581055,0.13333332538604736,0.8947917222976685,0.9722222089767456],"occluded":false},{"timestamp":13.52,"box":[0.4208333492279053,0.1379629671573639,0.8656250536441803,0.9731481373310089],"occluded":false},{"timestamp":13.780000000000001,"box":[0.35572919249534607,0.17592592537403107,0.7588542103767395,0.9722222238779068],"occluded":false},{"timestamp":14.040000000000001,"box":[0.2526041865348816,0.182407408952713,0.6328125298023224,0.972222238779068],"occluded":false},{"timestamp":14.3,"box":[0.1302083432674408,0.1527777761220932,0.5536458790302277,0.9722221940755844],"occluded":false},{"timestamp":14.56,"box":[0.09270834177732468,0.10462962836027145,0.5416667088866234,0.972222201526165],"occluded":false},{"timestamp":14.82,"box":[0.0,0.028703702613711357,0.4427083432674408,0.971296263858676],"occluded":false},{"timestamp":15.08,"box":[0.0,0.028703702613711357,0.13385418057441711,0.9749999772757292],"occluded":false}],"mean_score":0.9726027250289917}],"dropped":{},"fates":[{"noun_phrase":"hand","object_id":0,"frames":7,"mean_score":0.8827838897705078,"mean_area":0.24821842660494225,"outcome":"kept","presentation_id":1,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"woman","object_id":0,"frames":59,"mean_score":0.9726027250289917,"mean_area":0.3449651093368902,"outcome":"kept","presentation_id":2,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null}],"gates":{"min_track_frames":3,"min_mean_score":0.5,"merge_iou":0.8,"max_tracklets":12}}},"render":{"inputs":{"tracklets":2,"as_video":true,"fps":1.0,"max_frames":30},"outputs":{"media":"/home/zzk/data/video_bbox/vlm_gemini3.5_lite/renders/--cif7Nv6BI_chunk_014_scene_013_choices.mp4","media_bytes":5679364,"frames":[{"index":1,"timestamp":0.0,"file":"--cif7Nv6BI_chunk_014_scene_013_choices.mp4"},{"index":2,"timestamp":1.04,"file":"--cif7Nv6BI_chunk_014_scene_013_choices.mp4"},{"index":3,"timestamp":2.08,"file":"--cif7Nv6BI_chunk_014_scene_013_choices.mp4"},{"index":4,"timestamp":2.86,"file":"--cif7Nv6BI_chunk_014_scene_013_choices.mp4"},{"index":5,"timestamp":3.9,"file":"--cif7Nv6BI_chunk_014_scene_013_choices.mp4"},{"index":6,"timestamp":4.94,"file":"--cif7Nv6BI_chunk_014_scene_013_choices.mp4"},{"index":7,"timestamp":5.98,"file":"--cif7Nv6BI_chunk_014_scene_013_choices.mp4"},{"index":8,"timestamp":6.76,"file":"--cif7Nv6BI_chunk_014_scene_013_choices.mp4"},{"index":9,"timestamp":7.8,"file":"--cif7Nv6BI_chunk_014_scene_013_choices.mp4"},{"index":10,"timestamp":8.84,"file":"--cif7Nv6BI_chunk_014_scene_013_choices.mp4"},{"index":11,"timestamp":9.88,"file":"--cif7Nv6BI_chunk_014_scene_013_choices.mp4"},{"index":12,"timestamp":10.92,"file":"--cif7Nv6BI_chunk_014_scene_013_choices.mp4"},{"index":13,"timestamp":11.700000000000001,"file":"--cif7Nv6BI_chunk_014_scene_013_choices.mp4"},{"index":14,"timestamp":12.74,"file":"--cif7Nv6BI_chunk_014_scene_013_choices.mp4"},{"index":15,"timestamp":13.780000000000001,"file":"--cif7Nv6BI_chunk_014_scene_013_choices.mp4"},{"index":16,"timestamp":14.82,"file":"--cif7Nv6BI_chunk_014_scene_013_choices.mp4"},{"index":17,"timestamp":15.6,"file":"--cif7Nv6BI_chunk_014_scene_013_choices.mp4"},{"index":18,"timestamp":16.64,"file":"--cif7Nv6BI_chunk_014_scene_013_choices.mp4"},{"index":19,"timestamp":17.68,"file":"--cif7Nv6BI_chunk_014_scene_013_choices.mp4"}]}},"select":{"inputs":{"model":"gemini-3.5-flash-lite","temperature":0.0,"as_video":true,"media_bytes":5679364,"media":["--cif7Nv6BI_chunk_014_scene_013_choices.mp4"],"choice_frames":19,"system_prompt":"You are labelling video data for referring multi-object tracking.\n\nYou will see a numbered sequence of frames from one clip, in time order. Each frame has its own number and timestamp printed in the top-left corner. Candidate objects have been detected for you and drawn on the frames as coloured boxes, each labelled with a number. The same object keeps the same number and the same colour on every frame. Below each frame is a strip of enlarged crops of those same numbered boxes, so you can identify a small or distant object.\n\nFor each description you are given, answer two questions:\n\n1. MATCHING_IDS — which of the numbered objects the description refers to. List every number that matches, or an EMPTY LIST if none of them do. An empty list is a correct and expected answer: some descriptions are deliberately about something that is not in this clip. Do not pick a number just to have picked one, and do not pick a number for an object that merely resembles the description.\n\n2. START_FRAME and END_FRAME — the frame numbers between which the described state actually holds. If the description is \"the car that's turning left\", the interval is the frames in which that car is turning, NOT every frame in which the car is visible. Use the frame numbers printed on the frames. START_FRAME must not be greater than END_FRAME. When MATCHING_IDS is empty, give the full range (the first and last frame numbers); it is ignored.\n\n3. CONFIDENCE — how sure you are of the MATCHING_IDS answer, from 0.0 to 1.0. Be honest: a low number costs the example nothing, and an overconfident wrong answer becomes a wrong label.\n\nRules:\n- Only use numbers that are actually drawn on the frames.\n- Never output coordinates, box values, or pixel positions.\n- Judge motion by comparing the object's position across frames, not by guessing from one frame.\n","user_prompt":"The video has 19 frames, numbered 1 to 19. Each frame is one moment of the clip and has its number burned into the top left corner.\n\nAnswer for each of the following descriptions, in the order given:\n1. the woman talking in front of the shop\n2. the hand waving across the camera view","questions":["the woman talking in front of the shop","the hand waving across the camera view"]},"outputs":{"raw_response":"{\n  \"answers\": [\n    {\n      \"description_index\": 1,\n      \"matching_ids\": [\n        2\n      ],\n      \"start_frame\": 1,\n      \"end_frame\": 16,\n      \"confidence\": 1.0\n    },\n    {\n      \"description_index\": 2,\n      \"matching_ids\": [\n        1\n      ],\n      \"start_frame\": 2,\n      \"end_frame\": 3,\n      \"confidence\": 1.0\n    }\n  ]\n}","raw_answers":[{"description_index":1,"matching_ids":[2],"start_frame":1,"end_frame":16,"confidence":1.0},{"description_index":2,"matching_ids":[1],"start_frame":2,"end_frame":3,"confidence":1.0}],"answers":[{"expression":"the woman talking in front of the shop","matching_presentation_ids":[2],"start_choice_frame":1,"end_choice_frame":16,"confidence":1.0,"self_reported_confidence":1.0},{"expression":"the hand waving across the camera view","matching_presentation_ids":[1],"start_choice_frame":2,"end_choice_frame":3,"confidence":1.0,"self_reported_confidence":1.0}]}},"verify":{"inputs":{"expressions":2,"tracklets":2,"selections":2,"gates":"interval, count, unknown ids, confidence, then geometry (motion)"},"outputs":{"kept":[{"expression":"the woman talking in front of the shop","kind":"appearance","is_negative":false,"tracks":[{"timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88,10.14,10.4,10.66,10.92,11.18,11.44,11.700000000000001,11.96,12.22,12.48,12.74,13.0,13.26,13.52,13.780000000000001,14.040000000000001,14.3,14.56,14.82],"boxes":[[0.2994791865348816,0.17129629850387573,0.7786458730697632,0.9712963104248047],[0.30937501788139343,0.1787036955356598,0.7463541924953461,0.9722221791744232],[0.2942708432674408,0.182407408952713,0.7125000357627869,0.9731481373310089],[0.3135416805744171,0.19814814627170563,0.7161458730697632,0.9740740805864334],[0.11562500894069672,0.23148147761821747,0.7416667193174362,0.974074050784111],[0.37916669249534607,0.260185182094574,0.7546875476837158,0.9731481075286865],[0.19947917759418488,0.287962943315506,0.7682292014360428,0.9731481373310089],[0.22708334028720856,0.2657407224178314,0.7718750387430191,0.9722221791744232],[0.40781253576278687,0.2657407224178314,0.7843750417232513,0.9722221791744232],[0.4140625298023224,0.2481481432914734,0.8000000417232513,0.9722222089767456],[0.17135417461395264,0.22962962090969086,0.8088542222976685,0.9712962955236435],[0.41822919249534607,0.21944443881511688,0.8083333671092987,0.9722221940755844],[0.3916666805744171,0.20370370149612427,0.7895833551883698,0.9722222089767456],[0.37291669845581055,0.18703703582286835,0.7848958969116211,0.9722221940755844],[0.37135419249534607,0.16018518805503845,0.7901042103767395,0.9712962806224823],[0.36250001192092896,0.15648147463798523,0.7791666984558105,0.9731481373310089],[0.3265625238418579,0.17222221195697784,0.7406250536441803,0.9740740209817886],[0.31197917461395264,0.1935185194015503,0.7145833671092987,0.9740740656852722],[0.2984375059604645,0.19907407462596893,0.6927083730697632,0.9731481522321701],[0.2848958373069763,0.18518517911434174,0.679166704416275,0.9731481522321701],[0.28177085518836975,0.1814814805984497,0.6770833730697632,0.9740740656852722],[0.2760416865348816,0.18703703582286835,0.6755208671092987,0.9750000089406967],[0.2578125,0.19074073433876038,0.6744791865348816,0.9740740358829498],[0.26979169249534607,0.19629628956317902,0.6817708909511566,0.9759258776903152],[0.2828125059604645,0.192592591047287,0.694791704416275,0.9759258925914764],[0.2968750298023224,0.17685185372829437,0.7161458730697632,0.9722221940755844],[0.30156251788139343,0.1657407432794571,0.7281250357627869,0.9731481224298477],[0.29479166865348816,0.14999999105930328,0.7270833551883698,0.9722222238779068],[0.2942708432674408,0.14351852238178253,0.7364583611488342,0.9722222238779068],[0.30156251788139343,0.13518518209457397,0.7473958730697632,0.9731481075286865],[0.3083333373069763,0.14999999105930328,0.7562500238418579,0.9731481224298477],[0.30781251192092896,0.1509259194135666,0.7520833611488342,0.9722221940755844],[0.30781251192092896,0.1509259194135666,0.7468750476837158,0.9731481522321701],[0.29635417461395264,0.13611111044883728,0.7437500357627869,0.9722221791744232],[0.2942708432674408,0.138888880610466,0.7385416924953461,0.9722221940755844],[0.29114586114883423,0.14351852238178253,0.7312500476837158,0.9731481224298477],[0.2786458432674408,0.14537036418914795,0.7151041924953461,0.9731481075286865],[0.2734375,0.15555554628372192,0.7026041746139526,0.9722222089767456],[0.2614583373069763,0.1546296328306198,0.6895833611488342,0.9731481522321701],[0.2526041865348816,0.1537037044763565,0.6786458790302277,0.9731481224298477],[0.25052085518836975,0.14537036418914795,0.6812500357627869,0.9722222089767456],[0.24687501788139343,0.16111110150814056,0.6760416924953461,0.9777777642011642],[0.2473958432674408,0.18333333730697632,0.6802083551883698,0.9740740656852722],[0.24635417759418488,0.16296295821666718,0.6833333522081375,0.974074050784111],[0.2343750149011612,0.14814814925193787,0.680208370089531,0.9749999940395355],[0.26875001192092896,0.16388888657093048,0.7161458730697632,0.9731481224298477],[0.2671875059604645,0.1787036955356598,0.7140625417232513,0.9740740358829498],[0.2723958492279053,0.17685185372829437,0.7161458730697632,0.9731481522321701],[0.2848958373069763,0.182407408952713,0.7265625298023224,0.9740740358829498],[0.3187500238418579,0.18981480598449707,0.7656250596046448,0.9740740656852722],[0.3609375059604645,0.1527777761220932,0.8343750238418579,0.9712962955236435],[0.41979169845581055,0.13333332538604736,0.8947917222976685,0.9722222089767456],[0.4208333492279053,0.1379629671573639,0.8656250536441803,0.9731481373310089],[0.35572919249534607,0.17592592537403107,0.7588542103767395,0.9722222238779068],[0.2526041865348816,0.182407408952713,0.6328125298023224,0.972222238779068],[0.1302083432674408,0.1527777761220932,0.5536458790302277,0.9722221940755844],[0.09270834177732468,0.10462962836027145,0.5416667088866234,0.972222201526165],[0.0,0.028703702613711357,0.4427083432674408,0.971296263858676]]}]},{"expression":"the hand waving across the camera view","kind":"motion","is_negative":false,"tracks":[{"timestamps":[1.04,1.3,1.56,1.82,2.08],"boxes":[[0.05677083507180214,0.7268518209457397,0.22500001266598701,0.9712962657213211],[0.04114583507180214,0.5009258985519409,0.37031251564621925,0.9712962508201599],[0.0937500074505806,0.45370370149612427,0.7411458566784859,0.9712963104248047],[0.09947916865348816,0.4611110985279083,0.5604166984558105,0.9722221791744232],[0.06093750149011612,0.3898147940635681,0.7583333626389503,0.9712962508201599]]}]}],"rejected":{},"rejection_counts":{},"row":{"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_014_scene_013.mp4","scene_key":"--cif7Nv6BI_chunk_014_scene_013","category":"test_vehicles","height":1080,"width":1920,"fps":50.0,"num_frames":893,"duration":17.86,"rmot_expressions":[{"expression":"the woman talking in front of the shop","kind":"appearance","is_negative":false,"tracks":[{"timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88,10.14,10.4,10.66,10.92,11.18,11.44,11.700000000000001,11.96,12.22,12.48,12.74,13.0,13.26,13.52,13.780000000000001,14.040000000000001,14.3,14.56,14.82],"boxes":[[0.2994791865348816,0.17129629850387573,0.7786458730697632,0.9712963104248047],[0.30937501788139343,0.1787036955356598,0.7463541924953461,0.9722221791744232],[0.2942708432674408,0.182407408952713,0.7125000357627869,0.9731481373310089],[0.3135416805744171,0.19814814627170563,0.7161458730697632,0.9740740805864334],[0.11562500894069672,0.23148147761821747,0.7416667193174362,0.974074050784111],[0.37916669249534607,0.260185182094574,0.7546875476837158,0.9731481075286865],[0.19947917759418488,0.287962943315506,0.7682292014360428,0.9731481373310089],[0.22708334028720856,0.2657407224178314,0.7718750387430191,0.9722221791744232],[0.40781253576278687,0.2657407224178314,0.7843750417232513,0.9722221791744232],[0.4140625298023224,0.2481481432914734,0.8000000417232513,0.9722222089767456],[0.17135417461395264,0.22962962090969086,0.8088542222976685,0.9712962955236435],[0.41822919249534607,0.21944443881511688,0.8083333671092987,0.9722221940755844],[0.3916666805744171,0.20370370149612427,0.7895833551883698,0.9722222089767456],[0.37291669845581055,0.18703703582286835,0.7848958969116211,0.9722221940755844],[0.37135419249534607,0.16018518805503845,0.7901042103767395,0.9712962806224823],[0.36250001192092896,0.15648147463798523,0.7791666984558105,0.9731481373310089],[0.3265625238418579,0.17222221195697784,0.7406250536441803,0.9740740209817886],[0.31197917461395264,0.1935185194015503,0.7145833671092987,0.9740740656852722],[0.2984375059604645,0.19907407462596893,0.6927083730697632,0.9731481522321701],[0.2848958373069763,0.18518517911434174,0.679166704416275,0.9731481522321701],[0.28177085518836975,0.1814814805984497,0.6770833730697632,0.9740740656852722],[0.2760416865348816,0.18703703582286835,0.6755208671092987,0.9750000089406967],[0.2578125,0.19074073433876038,0.6744791865348816,0.9740740358829498],[0.26979169249534607,0.19629628956317902,0.6817708909511566,0.9759258776903152],[0.2828125059604645,0.192592591047287,0.694791704416275,0.9759258925914764],[0.2968750298023224,0.17685185372829437,0.7161458730697632,0.9722221940755844],[0.30156251788139343,0.1657407432794571,0.7281250357627869,0.9731481224298477],[0.29479166865348816,0.14999999105930328,0.7270833551883698,0.9722222238779068],[0.2942708432674408,0.14351852238178253,0.7364583611488342,0.9722222238779068],[0.30156251788139343,0.13518518209457397,0.7473958730697632,0.9731481075286865],[0.3083333373069763,0.14999999105930328,0.7562500238418579,0.9731481224298477],[0.30781251192092896,0.1509259194135666,0.7520833611488342,0.9722221940755844],[0.30781251192092896,0.1509259194135666,0.7468750476837158,0.9731481522321701],[0.29635417461395264,0.13611111044883728,0.7437500357627869,0.9722221791744232],[0.2942708432674408,0.138888880610466,0.7385416924953461,0.9722221940755844],[0.29114586114883423,0.14351852238178253,0.7312500476837158,0.9731481224298477],[0.2786458432674408,0.14537036418914795,0.7151041924953461,0.9731481075286865],[0.2734375,0.15555554628372192,0.7026041746139526,0.9722222089767456],[0.2614583373069763,0.1546296328306198,0.6895833611488342,0.9731481522321701],[0.2526041865348816,0.1537037044763565,0.6786458790302277,0.9731481224298477],[0.25052085518836975,0.14537036418914795,0.6812500357627869,0.9722222089767456],[0.24687501788139343,0.16111110150814056,0.6760416924953461,0.9777777642011642],[0.2473958432674408,0.18333333730697632,0.6802083551883698,0.9740740656852722],[0.24635417759418488,0.16296295821666718,0.6833333522081375,0.974074050784111],[0.2343750149011612,0.14814814925193787,0.680208370089531,0.9749999940395355],[0.26875001192092896,0.16388888657093048,0.7161458730697632,0.9731481224298477],[0.2671875059604645,0.1787036955356598,0.7140625417232513,0.9740740358829498],[0.2723958492279053,0.17685185372829437,0.7161458730697632,0.9731481522321701],[0.2848958373069763,0.182407408952713,0.7265625298023224,0.9740740358829498],[0.3187500238418579,0.18981480598449707,0.7656250596046448,0.9740740656852722],[0.3609375059604645,0.1527777761220932,0.8343750238418579,0.9712962955236435],[0.41979169845581055,0.13333332538604736,0.8947917222976685,0.9722222089767456],[0.4208333492279053,0.1379629671573639,0.8656250536441803,0.9731481373310089],[0.35572919249534607,0.17592592537403107,0.7588542103767395,0.9722222238779068],[0.2526041865348816,0.182407408952713,0.6328125298023224,0.972222238779068],[0.1302083432674408,0.1527777761220932,0.5536458790302277,0.9722221940755844],[0.09270834177732468,0.10462962836027145,0.5416667088866234,0.972222201526165],[0.0,0.028703702613711357,0.4427083432674408,0.971296263858676]]}]},{"expression":"the hand waving across the camera view","kind":"motion","is_negative":false,"tracks":[{"timestamps":[1.04,1.3,1.56,1.82,2.08],"boxes":[[0.05677083507180214,0.7268518209457397,0.22500001266598701,0.9712962657213211],[0.04114583507180214,0.5009258985519409,0.37031251564621925,0.9712962508201599],[0.0937500074505806,0.45370370149612427,0.7411458566784859,0.9712963104248047],[0.09947916865348816,0.4611110985279083,0.5604166984558105,0.9722221791744232],[0.06093750149011612,0.3898147940635681,0.7583333626389503,0.9712962508201599]]}]}],"rmot_expression_count":2,"rmot_negative_count":0,"rmot_motion_count":1}}}}},{"index":8,"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--Dx9HVcQWs_chunk_002_scene_013.mp4","status":"complete","steps":{"decode":{"inputs":{"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--Dx9HVcQWs_chunk_002_scene_013.mp4","sample_fps":4.0},"outputs":{"width":1920,"height":1080,"source_fps":23.976023976023978,"duration_seconds":16.016,"frames_kept":65,"blank_frames_dropped":0,"timestamps":[0.0,0.25025,0.5005,0.75075,1.001,1.25125,1.5015,1.75175,2.002,2.25225,2.5025,2.75275,3.003,3.25325,3.5035,3.7537499999999997,4.004,4.25425,4.5045,4.75475,5.005,5.25525,5.5055,5.75575,6.006,6.25625,6.5065,6.756749999999999,7.007,7.25725,7.507499999999999,7.75775,8.008,8.25825,8.5085,8.75875,9.009,9.25925,9.5095,9.75975,10.01,10.26025,10.5105,10.76075,11.011,11.26125,11.5115,11.76175,12.012,12.26225,12.5125,12.762749999999999,13.013,13.26325,13.513499999999999,13.76375,14.014,14.264249999999999,14.5145,14.76475,15.014999999999999,15.26525,15.5155,15.765749999999999,16.016],"error":""}},"propose":{"inputs":{"model":"gemini-3.5-flash-lite","temperature":0.4,"video_bytes":6122951,"system_prompt":"You are annotating video clips to build training data for referring multi-object tracking: given a video and a natural-language description, a model must output the boxes of every object matching that description, over time.\n\nYour job is to write the descriptions. You do NOT output coordinates, boxes, pixel positions, or any numbers about where things are. A separate detector finds the objects.\n\nFor every clip you produce a list of items. Each item has two texts that must NOT be the same:\n\nEXPRESSION — what a person would say to single out the object or objects. It may describe motion, temporal order, interaction with other objects, and spatial relations. It should read like something a person would actually say.\n  Good: \"the black car that's turning left\"\n  Good: \"the cyclist who overtakes the bus\"\n  Good: \"the two people who walk out of frame on the right\"\n\nNOUN_PHRASE — the same object as a plain, timeless visual category with its visible attributes, and NOTHING else. The detector is prompted with this text and looks at ONE frame at a time, so anything about time, order, or action is useless to it and makes it fail.\n  Allowed: category nouns, colours, materials, clothing, size, type\n  Forbidden: verbs and participles of action (\"turning\", \"walks\", \"is moving\"), temporal words (\"then\", \"starts\", \"after\", \"first\", \"last\"), relative clauses (\"that ...\", \"which ...\", \"who ...\"), ordinals and rank words (\"first\", \"leftmost\", \"nearest\"), and anything over six words.\n  \"the black car that's turning left\" -> \"black car\"\n  \"the cyclist who overtakes the bus\" -> \"cyclist\"\n  \"the two people who walk out of frame on the right\" -> \"person\"\n\nKIND — what makes the expression hard: \"motion\", \"appearance\", \"spatial\", or \"interaction\". Aim for about 35% \"motion\" items: motion is the part a single-frame model cannot already solve, so it is the most valuable, but a list of only motion items is not usable training data either.\n\nEXPECTED_COUNT — how many distinct objects in the clip match the EXPRESSION. Count carefully; this is checked against the detector's answer, and a wrong count discards the item.\n\nIS_NEGATIVE — at most ONE item may be a negative, and only when you also write at least one non-negative item. A negative is an expression that recombines attributes that ARE present in the clip into a whole that is NOT. If the only black car is moving, \"the parked black car\" is a good negative. Its NOUN_PHRASE must still be the category that IS present (\"black car\"), because that is what makes the absence checkable rather than merely unverified. A negative's EXPECTED_COUNT is 0.\n\nRules that discard an item, so follow them exactly:\n- Never write coordinates, box values, pixel positions, or percentages of the frame.\n- Never let a NOUN_PHRASE contain a verb of action, a temporal word, a relative clause, or an ordinal.\n- Never write an EXPRESSION whose object is not visible in the clip, unless it is marked IS_NEGATIVE.\n- Prefer objects that are clearly visible and identifiable over background clutter.\n- Do not describe the camera or the shot; describe objects in the scene.\n","user_prompt":"Watch this clip and write between 1 and 3 items following the rules. Cover different objects rather than restating one object several ways, and include any negative in the same list.\n"},"outputs":{"raw_response":"{\n  \"items\": [\n    {\n      \"expression\": \"the man wearing a green hoodie and a baseball cap\",\n      \"noun_phrase\": \"man\",\n      \"kind\": \"appearance\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    }\n  ]\n}","raw_items":[{"expression":"the man wearing a green hoodie and a baseball cap","noun_phrase":"man","kind":"appearance","expected_count":1,"is_negative":false}],"accepted":[{"expression":"the man wearing a green hoodie and a baseball cap","noun_phrase":"man","kind":"appearance","expected_count":1,"is_negative":false}],"gated":[]}},"track":{"inputs":{"tracker":"Sam31MultiplexTracker","model_id":"/mnt/ceph3/ec/jacky/sam3.1","noun_phrases":["man"],"expressions_in":1,"frames":65,"sample_fps":4.0,"frame_size":"1920x1080","timestamps":[0.0,0.25025,0.5005,0.75075,1.001,1.25125,1.5015,1.75175,2.002,2.25225,2.5025,2.75275,3.003,3.25325,3.5035,3.7537499999999997,4.004,4.25425,4.5045,4.75475,5.005,5.25525,5.5055,5.75575,6.006,6.25625,6.5065,6.756749999999999,7.007,7.25725,7.507499999999999,7.75775,8.008,8.25825,8.5085,8.75875,9.009,9.25925,9.5095,9.75975,10.01,10.26025,10.5105,10.76075,11.011,11.26125,11.5115,11.76175,12.012,12.26225,12.5125,12.762749999999999,13.013,13.26325,13.513499999999999,13.76375,14.014,14.264249999999999,14.5145,14.76475,15.014999999999999,15.26525,15.5155,15.765749999999999,16.016],"session":"one session over 65 JPEG frames; then per phrase: reset_session, add_prompt(text) on frame 0, propagate_in_video","passes":1},"outputs":{"raw_tracks":[{"noun_phrase":"man","object_id":0,"observations":[{"timestamp":0.0,"box":[0.10572917014360428,0.017592592164874077,0.9020834043622017,0.9990740846842527],"occluded":false},{"timestamp":0.25025,"box":[0.11666667461395264,0.01666666567325592,0.9203125238418579,0.9990740567445755],"occluded":false},{"timestamp":0.5005,"box":[0.17395834624767303,0.024074073880910873,0.9630209058523178,0.999074038118124],"occluded":false},{"timestamp":0.75075,"box":[0.16979166865348816,0.024074073880910873,0.9031250178813934,0.999074038118124],"occluded":false},{"timestamp":1.001,"box":[0.16562500596046448,0.024074073880910873,0.8859375417232513,0.999074038118124],"occluded":false},{"timestamp":1.25125,"box":[0.16197916865348816,0.023148147389292717,0.8828125298023224,0.9990740697830915],"occluded":false},{"timestamp":1.5015,"box":[0.15833334624767303,0.014814814552664757,0.8921875804662704,0.999074062332511],"occluded":false},{"timestamp":1.75175,"box":[0.1588541716337204,0.01944444328546524,0.8697917312383652,0.9990740790963173],"occluded":false},{"timestamp":2.002,"box":[0.15000000596046448,0.024074073880910873,0.8927083909511566,0.999074038118124],"occluded":false},{"timestamp":2.25225,"box":[0.15468750894069672,0.020370369777083397,0.8984375447034836,0.9990740474313498],"occluded":false},{"timestamp":2.5025,"box":[0.1536458432674408,0.01944444328546524,0.8973958790302277,0.9990740790963173],"occluded":false},{"timestamp":2.75275,"box":[0.15208333730697632,0.02222222089767456,0.8937500715255737,0.9990740418434143],"occluded":false},{"timestamp":3.003,"box":[0.15677084028720856,0.01944444328546524,0.8937500268220901,0.9990740790963173],"occluded":false},{"timestamp":3.25325,"box":[0.16041667759418488,0.012037036940455437,0.8927083760499954,0.9990740399807692],"occluded":false},{"timestamp":3.5035,"box":[0.1614583432674408,0.012962962500751019,0.8968750536441803,0.9990740669891238],"occluded":false},{"timestamp":3.7537499999999997,"box":[0.16093750298023224,0.017592592164874077,0.8927083760499954,0.9990740846842527],"occluded":false},{"timestamp":4.004,"box":[0.1614583432674408,0.028703702613711357,0.8895833790302277,0.9990740548819304],"occluded":false},{"timestamp":4.25425,"box":[0.1588541716337204,0.02500000037252903,0.8932292312383652,0.9990740660578012],"occluded":false},{"timestamp":4.5045,"box":[0.15781250596046448,0.023148147389292717,0.893229216337204,0.9990740697830915],"occluded":false},{"timestamp":4.75475,"box":[0.15781250596046448,0.02500000037252903,0.8927083909511566,0.9990740660578012],"occluded":false},{"timestamp":5.005,"box":[0.1588541716337204,0.025925925001502037,0.8921875208616257,0.9990740325301886],"occluded":false},{"timestamp":5.25525,"box":[0.15781250596046448,0.02500000037252903,0.8921875655651093,0.9990740660578012],"occluded":false},{"timestamp":5.5055,"box":[0.15677084028720856,0.02500000037252903,0.8916667252779007,0.9990740660578012],"occluded":false},{"timestamp":5.75575,"box":[0.1562500149011612,0.02500000037252903,0.8942709118127823,0.9990740660578012],"occluded":false},{"timestamp":6.006,"box":[0.15937501192092896,0.018518518656492233,0.8932292461395264,0.9990740530192852],"occluded":false},{"timestamp":6.25625,"box":[0.12812501192092896,0.024074073880910873,0.9994792342185974,0.999074038118124],"occluded":false},{"timestamp":6.5065,"box":[0.011979167349636555,0.026851851493120193,0.9994792388752103,0.9990740604698658],"occluded":false},{"timestamp":6.756749999999999,"box":[0.0026041667442768812,0.02500000037252903,0.9994792144279927,0.9990740660578012],"occluded":false},{"timestamp":7.007,"box":[0.0182291679084301,0.023148147389292717,0.9994792155921459,0.9990740697830915],"occluded":false},{"timestamp":7.25725,"box":[0.15572917461395264,0.02500000037252903,0.9010416865348816,0.9990740660578012],"occluded":false},{"timestamp":7.507499999999999,"box":[0.15572917461395264,0.01944444328546524,0.9036458730697632,0.9990740790963173],"occluded":false},{"timestamp":7.75775,"box":[0.15677084028720856,0.018518518656492233,0.9026042371988297,0.9990740530192852],"occluded":false},{"timestamp":8.008,"box":[0.15833334624767303,0.012962962500751019,0.9000000804662704,0.9990740669891238],"occluded":false},{"timestamp":8.25825,"box":[0.15416666865348816,0.0,0.901041716337204,0.9990740418434143],"occluded":false},{"timestamp":8.5085,"box":[0.15468750894069672,0.0,0.9041667431592941,0.9990740418434143],"occluded":false},{"timestamp":8.75875,"box":[0.15572917461395264,0.0018518518190830946,0.9000000357627869,0.9990740369539708],"occluded":false},{"timestamp":9.009,"box":[0.15468750894069672,0.010185184888541698,0.8942708820104599,0.999074044637382],"occluded":false},{"timestamp":9.25925,"box":[0.15677084028720856,0.01944444328546524,0.8880208879709244,0.9990740790963173],"occluded":false},{"timestamp":9.5095,"box":[0.1562500149011612,0.024074073880910873,0.8869792371988297,0.999074038118124],"occluded":false},{"timestamp":9.75975,"box":[0.15520833432674408,0.02777777798473835,0.8859375566244125,0.9990740288048983],"occluded":false},{"timestamp":10.01,"box":[0.1536458432674408,0.024074073880910873,0.8859375417232513,0.999074038118124],"occluded":false},{"timestamp":10.26025,"box":[0.15416666865348816,0.024074073880910873,0.8859375417232513,0.999074038118124],"occluded":false},{"timestamp":10.5105,"box":[0.15468750894069672,0.025925925001502037,0.8854167312383652,0.9990740325301886],"occluded":false},{"timestamp":10.76075,"box":[0.15468750894069672,0.029629629105329514,0.8848959058523178,0.9990740828216076],"occluded":false},{"timestamp":11.011,"box":[0.15468750894069672,0.02222222089767456,0.8817708939313889,0.9990740418434143],"occluded":false},{"timestamp":11.26125,"box":[0.03697916865348816,0.029629629105329514,0.955729216337204,0.9990740828216076],"occluded":false},{"timestamp":11.5115,"box":[0.0520833358168602,0.03611110895872116,0.9416667446494102,0.9990740343928337],"occluded":false},{"timestamp":11.76175,"box":[0.15781250596046448,0.03888888657093048,0.8781250417232513,0.9990740567445755],"occluded":false},{"timestamp":12.012,"box":[0.15468750894069672,0.034259259700775146,0.8760416954755783,0.9990740418434143],"occluded":false},{"timestamp":12.26225,"box":[0.15520833432674408,0.034259259700775146,0.8812500685453415,0.9990740418434143],"occluded":false},{"timestamp":12.5125,"box":[0.15520833432674408,0.034259259700775146,0.8781250566244125,0.9990740418434143],"occluded":false},{"timestamp":12.762749999999999,"box":[0.15520833432674408,0.034259259700775146,0.8796875327825546,0.9990740418434143],"occluded":false},{"timestamp":13.013,"box":[0.1562500149011612,0.035185184329748154,0.8791667371988297,0.9990740679204464],"occluded":false},{"timestamp":13.26325,"box":[0.1562500149011612,0.03703703731298447,0.8786459118127823,0.9990740641951561],"occluded":false},{"timestamp":13.513499999999999,"box":[0.15677084028720856,0.037962961941957474,0.8781250268220901,0.9990740306675434],"occluded":false},{"timestamp":13.76375,"box":[0.15677084028720856,0.03703703731298447,0.8781250268220901,0.9990740641951561],"occluded":false},{"timestamp":14.014,"box":[0.15781250596046448,0.037962961941957474,0.877604216337204,0.9990740306675434],"occluded":false},{"timestamp":14.264249999999999,"box":[0.15833334624767303,0.0416666641831398,0.8781250566244125,0.9990740790963173],"occluded":false},{"timestamp":14.5145,"box":[0.15781250596046448,0.04259259253740311,0.8770833909511566,0.9990740492939949],"occluded":false},{"timestamp":14.76475,"box":[0.15781250596046448,0.0416666641831398,0.8765625655651093,0.9990740790963173],"occluded":false},{"timestamp":15.014999999999999,"box":[0.15833334624767303,0.04259259253740311,0.8765625208616257,0.9990740492939949],"occluded":false},{"timestamp":15.26525,"box":[0.15833334624767303,0.043518517166376114,0.8750000447034836,0.999074075371027],"occluded":false},{"timestamp":15.5155,"box":[0.15781250596046448,0.04259259253740311,0.8755208551883698,0.9990740492939949],"occluded":false},{"timestamp":15.765749999999999,"box":[0.15677084028720856,0.04259259253740311,0.8765625506639481,0.9990740492939949],"occluded":false},{"timestamp":16.016,"box":[0.15833334624767303,0.043518517166376114,0.8750000447034836,0.999074075371027],"occluded":false}],"scores":[0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917]}],"tracklets":[{"noun_phrase":"man","object_id":0,"presentation_id":1,"observations":[{"timestamp":0.0,"box":[0.10572917014360428,0.017592592164874077,0.9020834043622017,0.9990740846842527],"occluded":false},{"timestamp":0.25025,"box":[0.11666667461395264,0.01666666567325592,0.9203125238418579,0.9990740567445755],"occluded":false},{"timestamp":0.5005,"box":[0.17395834624767303,0.024074073880910873,0.9630209058523178,0.999074038118124],"occluded":false},{"timestamp":0.75075,"box":[0.16979166865348816,0.024074073880910873,0.9031250178813934,0.999074038118124],"occluded":false},{"timestamp":1.001,"box":[0.16562500596046448,0.024074073880910873,0.8859375417232513,0.999074038118124],"occluded":false},{"timestamp":1.25125,"box":[0.16197916865348816,0.023148147389292717,0.8828125298023224,0.9990740697830915],"occluded":false},{"timestamp":1.5015,"box":[0.15833334624767303,0.014814814552664757,0.8921875804662704,0.999074062332511],"occluded":false},{"timestamp":1.75175,"box":[0.1588541716337204,0.01944444328546524,0.8697917312383652,0.9990740790963173],"occluded":false},{"timestamp":2.002,"box":[0.15000000596046448,0.024074073880910873,0.8927083909511566,0.999074038118124],"occluded":false},{"timestamp":2.25225,"box":[0.15468750894069672,0.020370369777083397,0.8984375447034836,0.9990740474313498],"occluded":false},{"timestamp":2.5025,"box":[0.1536458432674408,0.01944444328546524,0.8973958790302277,0.9990740790963173],"occluded":false},{"timestamp":2.75275,"box":[0.15208333730697632,0.02222222089767456,0.8937500715255737,0.9990740418434143],"occluded":false},{"timestamp":3.003,"box":[0.15677084028720856,0.01944444328546524,0.8937500268220901,0.9990740790963173],"occluded":false},{"timestamp":3.25325,"box":[0.16041667759418488,0.012037036940455437,0.8927083760499954,0.9990740399807692],"occluded":false},{"timestamp":3.5035,"box":[0.1614583432674408,0.012962962500751019,0.8968750536441803,0.9990740669891238],"occluded":false},{"timestamp":3.7537499999999997,"box":[0.16093750298023224,0.017592592164874077,0.8927083760499954,0.9990740846842527],"occluded":false},{"timestamp":4.004,"box":[0.1614583432674408,0.028703702613711357,0.8895833790302277,0.9990740548819304],"occluded":false},{"timestamp":4.25425,"box":[0.1588541716337204,0.02500000037252903,0.8932292312383652,0.9990740660578012],"occluded":false},{"timestamp":4.5045,"box":[0.15781250596046448,0.023148147389292717,0.893229216337204,0.9990740697830915],"occluded":false},{"timestamp":4.75475,"box":[0.15781250596046448,0.02500000037252903,0.8927083909511566,0.9990740660578012],"occluded":false},{"timestamp":5.005,"box":[0.1588541716337204,0.025925925001502037,0.8921875208616257,0.9990740325301886],"occluded":false},{"timestamp":5.25525,"box":[0.15781250596046448,0.02500000037252903,0.8921875655651093,0.9990740660578012],"occluded":false},{"timestamp":5.5055,"box":[0.15677084028720856,0.02500000037252903,0.8916667252779007,0.9990740660578012],"occluded":false},{"timestamp":5.75575,"box":[0.1562500149011612,0.02500000037252903,0.8942709118127823,0.9990740660578012],"occluded":false},{"timestamp":6.006,"box":[0.15937501192092896,0.018518518656492233,0.8932292461395264,0.9990740530192852],"occluded":false},{"timestamp":6.25625,"box":[0.12812501192092896,0.024074073880910873,0.9994792342185974,0.999074038118124],"occluded":false},{"timestamp":6.5065,"box":[0.011979167349636555,0.026851851493120193,0.9994792388752103,0.9990740604698658],"occluded":false},{"timestamp":6.756749999999999,"box":[0.0026041667442768812,0.02500000037252903,0.9994792144279927,0.9990740660578012],"occluded":false},{"timestamp":7.007,"box":[0.0182291679084301,0.023148147389292717,0.9994792155921459,0.9990740697830915],"occluded":false},{"timestamp":7.25725,"box":[0.15572917461395264,0.02500000037252903,0.9010416865348816,0.9990740660578012],"occluded":false},{"timestamp":7.507499999999999,"box":[0.15572917461395264,0.01944444328546524,0.9036458730697632,0.9990740790963173],"occluded":false},{"timestamp":7.75775,"box":[0.15677084028720856,0.018518518656492233,0.9026042371988297,0.9990740530192852],"occluded":false},{"timestamp":8.008,"box":[0.15833334624767303,0.012962962500751019,0.9000000804662704,0.9990740669891238],"occluded":false},{"timestamp":8.25825,"box":[0.15416666865348816,0.0,0.901041716337204,0.9990740418434143],"occluded":false},{"timestamp":8.5085,"box":[0.15468750894069672,0.0,0.9041667431592941,0.9990740418434143],"occluded":false},{"timestamp":8.75875,"box":[0.15572917461395264,0.0018518518190830946,0.9000000357627869,0.9990740369539708],"occluded":false},{"timestamp":9.009,"box":[0.15468750894069672,0.010185184888541698,0.8942708820104599,0.999074044637382],"occluded":false},{"timestamp":9.25925,"box":[0.15677084028720856,0.01944444328546524,0.8880208879709244,0.9990740790963173],"occluded":false},{"timestamp":9.5095,"box":[0.1562500149011612,0.024074073880910873,0.8869792371988297,0.999074038118124],"occluded":false},{"timestamp":9.75975,"box":[0.15520833432674408,0.02777777798473835,0.8859375566244125,0.9990740288048983],"occluded":false},{"timestamp":10.01,"box":[0.1536458432674408,0.024074073880910873,0.8859375417232513,0.999074038118124],"occluded":false},{"timestamp":10.26025,"box":[0.15416666865348816,0.024074073880910873,0.8859375417232513,0.999074038118124],"occluded":false},{"timestamp":10.5105,"box":[0.15468750894069672,0.025925925001502037,0.8854167312383652,0.9990740325301886],"occluded":false},{"timestamp":10.76075,"box":[0.15468750894069672,0.029629629105329514,0.8848959058523178,0.9990740828216076],"occluded":false},{"timestamp":11.011,"box":[0.15468750894069672,0.02222222089767456,0.8817708939313889,0.9990740418434143],"occluded":false},{"timestamp":11.26125,"box":[0.03697916865348816,0.029629629105329514,0.955729216337204,0.9990740828216076],"occluded":false},{"timestamp":11.5115,"box":[0.0520833358168602,0.03611110895872116,0.9416667446494102,0.9990740343928337],"occluded":false},{"timestamp":11.76175,"box":[0.15781250596046448,0.03888888657093048,0.8781250417232513,0.9990740567445755],"occluded":false},{"timestamp":12.012,"box":[0.15468750894069672,0.034259259700775146,0.8760416954755783,0.9990740418434143],"occluded":false},{"timestamp":12.26225,"box":[0.15520833432674408,0.034259259700775146,0.8812500685453415,0.9990740418434143],"occluded":false},{"timestamp":12.5125,"box":[0.15520833432674408,0.034259259700775146,0.8781250566244125,0.9990740418434143],"occluded":false},{"timestamp":12.762749999999999,"box":[0.15520833432674408,0.034259259700775146,0.8796875327825546,0.9990740418434143],"occluded":false},{"timestamp":13.013,"box":[0.1562500149011612,0.035185184329748154,0.8791667371988297,0.9990740679204464],"occluded":false},{"timestamp":13.26325,"box":[0.1562500149011612,0.03703703731298447,0.8786459118127823,0.9990740641951561],"occluded":false},{"timestamp":13.513499999999999,"box":[0.15677084028720856,0.037962961941957474,0.8781250268220901,0.9990740306675434],"occluded":false},{"timestamp":13.76375,"box":[0.15677084028720856,0.03703703731298447,0.8781250268220901,0.9990740641951561],"occluded":false},{"timestamp":14.014,"box":[0.15781250596046448,0.037962961941957474,0.877604216337204,0.9990740306675434],"occluded":false},{"timestamp":14.264249999999999,"box":[0.15833334624767303,0.0416666641831398,0.8781250566244125,0.9990740790963173],"occluded":false},{"timestamp":14.5145,"box":[0.15781250596046448,0.04259259253740311,0.8770833909511566,0.9990740492939949],"occluded":false},{"timestamp":14.76475,"box":[0.15781250596046448,0.0416666641831398,0.8765625655651093,0.9990740790963173],"occluded":false},{"timestamp":15.014999999999999,"box":[0.15833334624767303,0.04259259253740311,0.8765625208616257,0.9990740492939949],"occluded":false},{"timestamp":15.26525,"box":[0.15833334624767303,0.043518517166376114,0.8750000447034836,0.999074075371027],"occluded":false},{"timestamp":15.5155,"box":[0.15781250596046448,0.04259259253740311,0.8755208551883698,0.9990740492939949],"occluded":false},{"timestamp":15.765749999999999,"box":[0.15677084028720856,0.04259259253740311,0.8765625506639481,0.9990740492939949],"occluded":false},{"timestamp":16.016,"box":[0.15833334624767303,0.043518517166376114,0.8750000447034836,0.999074075371027],"occluded":false}],"mean_score":0.9726027250289917}],"dropped":{},"fates":[{"noun_phrase":"man","object_id":0,"frames":65,"mean_score":0.9726027250289917,"mean_area":0.732847364166289,"outcome":"kept","presentation_id":1,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null}],"gates":{"min_track_frames":3,"min_mean_score":0.5,"merge_iou":0.8,"max_tracklets":12}}},"render":{"inputs":{"tracklets":1,"as_video":true,"fps":1.0,"max_frames":30},"outputs":{"media":"/home/zzk/data/video_bbox/vlm_gemini3.5_lite/renders/--Dx9HVcQWs_chunk_002_scene_013_choices.mp4","media_bytes":2522200,"frames":[{"index":1,"timestamp":0.0,"file":"--Dx9HVcQWs_chunk_002_scene_013_choices.mp4"},{"index":2,"timestamp":1.001,"file":"--Dx9HVcQWs_chunk_002_scene_013_choices.mp4"},{"index":3,"timestamp":2.002,"file":"--Dx9HVcQWs_chunk_002_scene_013_choices.mp4"},{"index":4,"timestamp":3.003,"file":"--Dx9HVcQWs_chunk_002_scene_013_choices.mp4"},{"index":5,"timestamp":4.004,"file":"--Dx9HVcQWs_chunk_002_scene_013_choices.mp4"},{"index":6,"timestamp":5.005,"file":"--Dx9HVcQWs_chunk_002_scene_013_choices.mp4"},{"index":7,"timestamp":6.006,"file":"--Dx9HVcQWs_chunk_002_scene_013_choices.mp4"},{"index":8,"timestamp":7.007,"file":"--Dx9HVcQWs_chunk_002_scene_013_choices.mp4"},{"index":9,"timestamp":8.008,"file":"--Dx9HVcQWs_chunk_002_scene_013_choices.mp4"},{"index":10,"timestamp":9.009,"file":"--Dx9HVcQWs_chunk_002_scene_013_choices.mp4"},{"index":11,"timestamp":10.01,"file":"--Dx9HVcQWs_chunk_002_scene_013_choices.mp4"},{"index":12,"timestamp":11.011,"file":"--Dx9HVcQWs_chunk_002_scene_013_choices.mp4"},{"index":13,"timestamp":12.012,"file":"--Dx9HVcQWs_chunk_002_scene_013_choices.mp4"},{"index":14,"timestamp":13.013,"file":"--Dx9HVcQWs_chunk_002_scene_013_choices.mp4"},{"index":15,"timestamp":14.014,"file":"--Dx9HVcQWs_chunk_002_scene_013_choices.mp4"},{"index":16,"timestamp":15.014999999999999,"file":"--Dx9HVcQWs_chunk_002_scene_013_choices.mp4"},{"index":17,"timestamp":16.016,"file":"--Dx9HVcQWs_chunk_002_scene_013_choices.mp4"}]}},"select":{"inputs":{"model":"gemini-3.5-flash-lite","temperature":0.0,"as_video":true,"media_bytes":2522200,"media":["--Dx9HVcQWs_chunk_002_scene_013_choices.mp4"],"choice_frames":17,"system_prompt":"You are labelling video data for referring multi-object tracking.\n\nYou will see a numbered sequence of frames from one clip, in time order. Each frame has its own number and timestamp printed in the top-left corner. Candidate objects have been detected for you and drawn on the frames as coloured boxes, each labelled with a number. The same object keeps the same number and the same colour on every frame. Below each frame is a strip of enlarged crops of those same numbered boxes, so you can identify a small or distant object.\n\nFor each description you are given, answer two questions:\n\n1. MATCHING_IDS — which of the numbered objects the description refers to. List every number that matches, or an EMPTY LIST if none of them do. An empty list is a correct and expected answer: some descriptions are deliberately about something that is not in this clip. Do not pick a number just to have picked one, and do not pick a number for an object that merely resembles the description.\n\n2. START_FRAME and END_FRAME — the frame numbers between which the described state actually holds. If the description is \"the car that's turning left\", the interval is the frames in which that car is turning, NOT every frame in which the car is visible. Use the frame numbers printed on the frames. START_FRAME must not be greater than END_FRAME. When MATCHING_IDS is empty, give the full range (the first and last frame numbers); it is ignored.\n\n3. CONFIDENCE — how sure you are of the MATCHING_IDS answer, from 0.0 to 1.0. Be honest: a low number costs the example nothing, and an overconfident wrong answer becomes a wrong label.\n\nRules:\n- Only use numbers that are actually drawn on the frames.\n- Never output coordinates, box values, or pixel positions.\n- Judge motion by comparing the object's position across frames, not by guessing from one frame.\n","user_prompt":"The video has 17 frames, numbered 1 to 17. Each frame is one moment of the clip and has its number burned into the top left corner.\n\nAnswer for each of the following descriptions, in the order given:\n1. the man wearing a green hoodie and a baseball cap","questions":["the man wearing a green hoodie and a baseball cap"]},"outputs":{"raw_response":"{\n  \"answers\": [\n    {\n      \"description_index\": 1,\n      \"matching_ids\": [\n        1\n      ],\n      \"start_frame\": 1,\n      \"end_frame\": 17,\n      \"confidence\": 1.0\n    }\n  ]\n}","raw_answers":[{"description_index":1,"matching_ids":[1],"start_frame":1,"end_frame":17,"confidence":1.0}],"answers":[{"expression":"the man wearing a green hoodie and a baseball cap","matching_presentation_ids":[1],"start_choice_frame":1,"end_choice_frame":17,"confidence":1.0,"self_reported_confidence":1.0}]}},"verify":{"inputs":{"expressions":1,"tracklets":1,"selections":1,"gates":"interval, count, unknown ids, confidence, then geometry (motion)"},"outputs":{"kept":[{"expression":"the man wearing a green hoodie and a baseball cap","kind":"appearance","is_negative":false,"tracks":[{"timestamps":[0.0,0.25025,0.5005,0.75075,1.001,1.25125,1.5015,1.75175,2.002,2.25225,2.5025,2.75275,3.003,3.25325,3.5035,3.7537499999999997,4.004,4.25425,4.5045,4.75475,5.005,5.25525,5.5055,5.75575,6.006,6.25625,6.5065,6.756749999999999,7.007,7.25725,7.507499999999999,7.75775,8.008,8.25825,8.5085,8.75875,9.009,9.25925,9.5095,9.75975,10.01,10.26025,10.5105,10.76075,11.011,11.26125,11.5115,11.76175,12.012,12.26225,12.5125,12.762749999999999,13.013,13.26325,13.513499999999999,13.76375,14.014,14.264249999999999,14.5145,14.76475,15.014999999999999,15.26525,15.5155,15.765749999999999,16.016],"boxes":[[0.10572917014360428,0.017592592164874077,0.9020834043622017,0.9990740846842527],[0.11666667461395264,0.01666666567325592,0.9203125238418579,0.9990740567445755],[0.17395834624767303,0.024074073880910873,0.9630209058523178,0.999074038118124],[0.16979166865348816,0.024074073880910873,0.9031250178813934,0.999074038118124],[0.16562500596046448,0.024074073880910873,0.8859375417232513,0.999074038118124],[0.16197916865348816,0.023148147389292717,0.8828125298023224,0.9990740697830915],[0.15833334624767303,0.014814814552664757,0.8921875804662704,0.999074062332511],[0.1588541716337204,0.01944444328546524,0.8697917312383652,0.9990740790963173],[0.15000000596046448,0.024074073880910873,0.8927083909511566,0.999074038118124],[0.15468750894069672,0.020370369777083397,0.8984375447034836,0.9990740474313498],[0.1536458432674408,0.01944444328546524,0.8973958790302277,0.9990740790963173],[0.15208333730697632,0.02222222089767456,0.8937500715255737,0.9990740418434143],[0.15677084028720856,0.01944444328546524,0.8937500268220901,0.9990740790963173],[0.16041667759418488,0.012037036940455437,0.8927083760499954,0.9990740399807692],[0.1614583432674408,0.012962962500751019,0.8968750536441803,0.9990740669891238],[0.16093750298023224,0.017592592164874077,0.8927083760499954,0.9990740846842527],[0.1614583432674408,0.028703702613711357,0.8895833790302277,0.9990740548819304],[0.1588541716337204,0.02500000037252903,0.8932292312383652,0.9990740660578012],[0.15781250596046448,0.023148147389292717,0.893229216337204,0.9990740697830915],[0.15781250596046448,0.02500000037252903,0.8927083909511566,0.9990740660578012],[0.1588541716337204,0.025925925001502037,0.8921875208616257,0.9990740325301886],[0.15781250596046448,0.02500000037252903,0.8921875655651093,0.9990740660578012],[0.15677084028720856,0.02500000037252903,0.8916667252779007,0.9990740660578012],[0.1562500149011612,0.02500000037252903,0.8942709118127823,0.9990740660578012],[0.15937501192092896,0.018518518656492233,0.8932292461395264,0.9990740530192852],[0.12812501192092896,0.024074073880910873,0.9994792342185974,0.999074038118124],[0.011979167349636555,0.026851851493120193,0.9994792388752103,0.9990740604698658],[0.0026041667442768812,0.02500000037252903,0.9994792144279927,0.9990740660578012],[0.0182291679084301,0.023148147389292717,0.9994792155921459,0.9990740697830915],[0.15572917461395264,0.02500000037252903,0.9010416865348816,0.9990740660578012],[0.15572917461395264,0.01944444328546524,0.9036458730697632,0.9990740790963173],[0.15677084028720856,0.018518518656492233,0.9026042371988297,0.9990740530192852],[0.15833334624767303,0.012962962500751019,0.9000000804662704,0.9990740669891238],[0.15416666865348816,0.0,0.901041716337204,0.9990740418434143],[0.15468750894069672,0.0,0.9041667431592941,0.9990740418434143],[0.15572917461395264,0.0018518518190830946,0.9000000357627869,0.9990740369539708],[0.15468750894069672,0.010185184888541698,0.8942708820104599,0.999074044637382],[0.15677084028720856,0.01944444328546524,0.8880208879709244,0.9990740790963173],[0.1562500149011612,0.024074073880910873,0.8869792371988297,0.999074038118124],[0.15520833432674408,0.02777777798473835,0.8859375566244125,0.9990740288048983],[0.1536458432674408,0.024074073880910873,0.8859375417232513,0.999074038118124],[0.15416666865348816,0.024074073880910873,0.8859375417232513,0.999074038118124],[0.15468750894069672,0.025925925001502037,0.8854167312383652,0.9990740325301886],[0.15468750894069672,0.029629629105329514,0.8848959058523178,0.9990740828216076],[0.15468750894069672,0.02222222089767456,0.8817708939313889,0.9990740418434143],[0.03697916865348816,0.029629629105329514,0.955729216337204,0.9990740828216076],[0.0520833358168602,0.03611110895872116,0.9416667446494102,0.9990740343928337],[0.15781250596046448,0.03888888657093048,0.8781250417232513,0.9990740567445755],[0.15468750894069672,0.034259259700775146,0.8760416954755783,0.9990740418434143],[0.15520833432674408,0.034259259700775146,0.8812500685453415,0.9990740418434143],[0.15520833432674408,0.034259259700775146,0.8781250566244125,0.9990740418434143],[0.15520833432674408,0.034259259700775146,0.8796875327825546,0.9990740418434143],[0.1562500149011612,0.035185184329748154,0.8791667371988297,0.9990740679204464],[0.1562500149011612,0.03703703731298447,0.8786459118127823,0.9990740641951561],[0.15677084028720856,0.037962961941957474,0.8781250268220901,0.9990740306675434],[0.15677084028720856,0.03703703731298447,0.8781250268220901,0.9990740641951561],[0.15781250596046448,0.037962961941957474,0.877604216337204,0.9990740306675434],[0.15833334624767303,0.0416666641831398,0.8781250566244125,0.9990740790963173],[0.15781250596046448,0.04259259253740311,0.8770833909511566,0.9990740492939949],[0.15781250596046448,0.0416666641831398,0.8765625655651093,0.9990740790963173],[0.15833334624767303,0.04259259253740311,0.8765625208616257,0.9990740492939949],[0.15833334624767303,0.043518517166376114,0.8750000447034836,0.999074075371027],[0.15781250596046448,0.04259259253740311,0.8755208551883698,0.9990740492939949],[0.15677084028720856,0.04259259253740311,0.8765625506639481,0.9990740492939949],[0.15833334624767303,0.043518517166376114,0.8750000447034836,0.999074075371027]]}]}],"rejected":{},"rejection_counts":{},"row":{"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--Dx9HVcQWs_chunk_002_scene_013.mp4","scene_key":"--Dx9HVcQWs_chunk_002_scene_013","category":"test_vehicles","height":1080,"width":1920,"fps":23.976023976023978,"num_frames":384,"duration":16.016,"rmot_expressions":[{"expression":"the man wearing a green hoodie and a baseball cap","kind":"appearance","is_negative":false,"tracks":[{"timestamps":[0.0,0.25025,0.5005,0.75075,1.001,1.25125,1.5015,1.75175,2.002,2.25225,2.5025,2.75275,3.003,3.25325,3.5035,3.7537499999999997,4.004,4.25425,4.5045,4.75475,5.005,5.25525,5.5055,5.75575,6.006,6.25625,6.5065,6.756749999999999,7.007,7.25725,7.507499999999999,7.75775,8.008,8.25825,8.5085,8.75875,9.009,9.25925,9.5095,9.75975,10.01,10.26025,10.5105,10.76075,11.011,11.26125,11.5115,11.76175,12.012,12.26225,12.5125,12.762749999999999,13.013,13.26325,13.513499999999999,13.76375,14.014,14.264249999999999,14.5145,14.76475,15.014999999999999,15.26525,15.5155,15.765749999999999,16.016],"boxes":[[0.10572917014360428,0.017592592164874077,0.9020834043622017,0.9990740846842527],[0.11666667461395264,0.01666666567325592,0.9203125238418579,0.9990740567445755],[0.17395834624767303,0.024074073880910873,0.9630209058523178,0.999074038118124],[0.16979166865348816,0.024074073880910873,0.9031250178813934,0.999074038118124],[0.16562500596046448,0.024074073880910873,0.8859375417232513,0.999074038118124],[0.16197916865348816,0.023148147389292717,0.8828125298023224,0.9990740697830915],[0.15833334624767303,0.014814814552664757,0.8921875804662704,0.999074062332511],[0.1588541716337204,0.01944444328546524,0.8697917312383652,0.9990740790963173],[0.15000000596046448,0.024074073880910873,0.8927083909511566,0.999074038118124],[0.15468750894069672,0.020370369777083397,0.8984375447034836,0.9990740474313498],[0.1536458432674408,0.01944444328546524,0.8973958790302277,0.9990740790963173],[0.15208333730697632,0.02222222089767456,0.8937500715255737,0.9990740418434143],[0.15677084028720856,0.01944444328546524,0.8937500268220901,0.9990740790963173],[0.16041667759418488,0.012037036940455437,0.8927083760499954,0.9990740399807692],[0.1614583432674408,0.012962962500751019,0.8968750536441803,0.9990740669891238],[0.16093750298023224,0.017592592164874077,0.8927083760499954,0.9990740846842527],[0.1614583432674408,0.028703702613711357,0.8895833790302277,0.9990740548819304],[0.1588541716337204,0.02500000037252903,0.8932292312383652,0.9990740660578012],[0.15781250596046448,0.023148147389292717,0.893229216337204,0.9990740697830915],[0.15781250596046448,0.02500000037252903,0.8927083909511566,0.9990740660578012],[0.1588541716337204,0.025925925001502037,0.8921875208616257,0.9990740325301886],[0.15781250596046448,0.02500000037252903,0.8921875655651093,0.9990740660578012],[0.15677084028720856,0.02500000037252903,0.8916667252779007,0.9990740660578012],[0.1562500149011612,0.02500000037252903,0.8942709118127823,0.9990740660578012],[0.15937501192092896,0.018518518656492233,0.8932292461395264,0.9990740530192852],[0.12812501192092896,0.024074073880910873,0.9994792342185974,0.999074038118124],[0.011979167349636555,0.026851851493120193,0.9994792388752103,0.9990740604698658],[0.0026041667442768812,0.02500000037252903,0.9994792144279927,0.9990740660578012],[0.0182291679084301,0.023148147389292717,0.9994792155921459,0.9990740697830915],[0.15572917461395264,0.02500000037252903,0.9010416865348816,0.9990740660578012],[0.15572917461395264,0.01944444328546524,0.9036458730697632,0.9990740790963173],[0.15677084028720856,0.018518518656492233,0.9026042371988297,0.9990740530192852],[0.15833334624767303,0.012962962500751019,0.9000000804662704,0.9990740669891238],[0.15416666865348816,0.0,0.901041716337204,0.9990740418434143],[0.15468750894069672,0.0,0.9041667431592941,0.9990740418434143],[0.15572917461395264,0.0018518518190830946,0.9000000357627869,0.9990740369539708],[0.15468750894069672,0.010185184888541698,0.8942708820104599,0.999074044637382],[0.15677084028720856,0.01944444328546524,0.8880208879709244,0.9990740790963173],[0.1562500149011612,0.024074073880910873,0.8869792371988297,0.999074038118124],[0.15520833432674408,0.02777777798473835,0.8859375566244125,0.9990740288048983],[0.1536458432674408,0.024074073880910873,0.8859375417232513,0.999074038118124],[0.15416666865348816,0.024074073880910873,0.8859375417232513,0.999074038118124],[0.15468750894069672,0.025925925001502037,0.8854167312383652,0.9990740325301886],[0.15468750894069672,0.029629629105329514,0.8848959058523178,0.9990740828216076],[0.15468750894069672,0.02222222089767456,0.8817708939313889,0.9990740418434143],[0.03697916865348816,0.029629629105329514,0.955729216337204,0.9990740828216076],[0.0520833358168602,0.03611110895872116,0.9416667446494102,0.9990740343928337],[0.15781250596046448,0.03888888657093048,0.8781250417232513,0.9990740567445755],[0.15468750894069672,0.034259259700775146,0.8760416954755783,0.9990740418434143],[0.15520833432674408,0.034259259700775146,0.8812500685453415,0.9990740418434143],[0.15520833432674408,0.034259259700775146,0.8781250566244125,0.9990740418434143],[0.15520833432674408,0.034259259700775146,0.8796875327825546,0.9990740418434143],[0.1562500149011612,0.035185184329748154,0.8791667371988297,0.9990740679204464],[0.1562500149011612,0.03703703731298447,0.8786459118127823,0.9990740641951561],[0.15677084028720856,0.037962961941957474,0.8781250268220901,0.9990740306675434],[0.15677084028720856,0.03703703731298447,0.8781250268220901,0.9990740641951561],[0.15781250596046448,0.037962961941957474,0.877604216337204,0.9990740306675434],[0.15833334624767303,0.0416666641831398,0.8781250566244125,0.9990740790963173],[0.15781250596046448,0.04259259253740311,0.8770833909511566,0.9990740492939949],[0.15781250596046448,0.0416666641831398,0.8765625655651093,0.9990740790963173],[0.15833334624767303,0.04259259253740311,0.8765625208616257,0.9990740492939949],[0.15833334624767303,0.043518517166376114,0.8750000447034836,0.999074075371027],[0.15781250596046448,0.04259259253740311,0.8755208551883698,0.9990740492939949],[0.15677084028720856,0.04259259253740311,0.8765625506639481,0.9990740492939949],[0.15833334624767303,0.043518517166376114,0.8750000447034836,0.999074075371027]]}]}],"rmot_expression_count":1,"rmot_negative_count":0,"rmot_motion_count":0}}}}},{"index":9,"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--Dx9HVcQWs_chunk_001_scene_034.mp4","status":"complete","steps":{"decode":{"inputs":{"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--Dx9HVcQWs_chunk_001_scene_034.mp4","sample_fps":4.0},"outputs":{"width":1920,"height":1080,"source_fps":23.976023976023978,"duration_seconds":14.097416666666666,"frames_kept":57,"blank_frames_dropped":0,"timestamps":[0.0,0.25025,0.5005,0.75075,1.001,1.25125,1.5015,1.75175,2.002,2.25225,2.5025,2.75275,3.003,3.25325,3.5035,3.7537499999999997,4.004,4.25425,4.5045,4.75475,5.005,5.25525,5.5055,5.75575,6.006,6.25625,6.5065,6.756749999999999,7.007,7.25725,7.507499999999999,7.75775,8.008,8.25825,8.5085,8.75875,9.009,9.25925,9.5095,9.75975,10.01,10.26025,10.5105,10.76075,11.011,11.26125,11.5115,11.76175,12.012,12.26225,12.5125,12.762749999999999,13.013,13.26325,13.513499999999999,13.76375,14.014],"error":""}},"propose":{"inputs":{"model":"gemini-3.5-flash-lite","temperature":0.4,"video_bytes":35175600,"system_prompt":"You are annotating video clips to build training data for referring multi-object tracking: given a video and a natural-language description, a model must output the boxes of every object matching that description, over time.\n\nYour job is to write the descriptions. You do NOT output coordinates, boxes, pixel positions, or any numbers about where things are. A separate detector finds the objects.\n\nFor every clip you produce a list of items. Each item has two texts that must NOT be the same:\n\nEXPRESSION — what a person would say to single out the object or objects. It may describe motion, temporal order, interaction with other objects, and spatial relations. It should read like something a person would actually say.\n  Good: \"the black car that's turning left\"\n  Good: \"the cyclist who overtakes the bus\"\n  Good: \"the two people who walk out of frame on the right\"\n\nNOUN_PHRASE — the same object as a plain, timeless visual category with its visible attributes, and NOTHING else. The detector is prompted with this text and looks at ONE frame at a time, so anything about time, order, or action is useless to it and makes it fail.\n  Allowed: category nouns, colours, materials, clothing, size, type\n  Forbidden: verbs and participles of action (\"turning\", \"walks\", \"is moving\"), temporal words (\"then\", \"starts\", \"after\", \"first\", \"last\"), relative clauses (\"that ...\", \"which ...\", \"who ...\"), ordinals and rank words (\"first\", \"leftmost\", \"nearest\"), and anything over six words.\n  \"the black car that's turning left\" -> \"black car\"\n  \"the cyclist who overtakes the bus\" -> \"cyclist\"\n  \"the two people who walk out of frame on the right\" -> \"person\"\n\nKIND — what makes the expression hard: \"motion\", \"appearance\", \"spatial\", or \"interaction\". Aim for about 35% \"motion\" items: motion is the part a single-frame model cannot already solve, so it is the most valuable, but a list of only motion items is not usable training data either.\n\nEXPECTED_COUNT — how many distinct objects in the clip match the EXPRESSION. Count carefully; this is checked against the detector's answer, and a wrong count discards the item.\n\nIS_NEGATIVE — at most ONE item may be a negative, and only when you also write at least one non-negative item. A negative is an expression that recombines attributes that ARE present in the clip into a whole that is NOT. If the only black car is moving, \"the parked black car\" is a good negative. Its NOUN_PHRASE must still be the category that IS present (\"black car\"), because that is what makes the absence checkable rather than merely unverified. A negative's EXPECTED_COUNT is 0.\n\nRules that discard an item, so follow them exactly:\n- Never write coordinates, box values, pixel positions, or percentages of the frame.\n- Never let a NOUN_PHRASE contain a verb of action, a temporal word, a relative clause, or an ordinal.\n- Never write an EXPRESSION whose object is not visible in the clip, unless it is marked IS_NEGATIVE.\n- Prefer objects that are clearly visible and identifiable over background clutter.\n- Do not describe the camera or the shot; describe objects in the scene.\n","user_prompt":"Watch this clip and write between 1 and 3 items following the rules. Cover different objects rather than restating one object several ways, and include any negative in the same list.\n"},"outputs":{"raw_response":"{\n  \"items\": [\n    {\n      \"expression\": \"the lead cyclist riding in front\",\n      \"noun_phrase\": \"cyclist in front\",\n      \"kind\": \"motion\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the cyclist wearing a bright pink jersey on the right\",\n      \"noun_phrase\": \"cyclist in pink jersey\",\n      \"kind\": \"appearance\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    }\n  ]\n}","raw_items":[{"expression":"the lead cyclist riding in front","noun_phrase":"cyclist in front","kind":"motion","expected_count":1,"is_negative":false},{"expression":"the cyclist wearing a bright pink jersey on the right","noun_phrase":"cyclist in pink jersey","kind":"appearance","expected_count":1,"is_negative":false}],"accepted":[{"expression":"the lead cyclist riding in front","noun_phrase":"cyclist in front","kind":"motion","expected_count":1,"is_negative":false},{"expression":"the cyclist wearing a bright pink jersey on the right","noun_phrase":"cyclist in pink jersey","kind":"appearance","expected_count":1,"is_negative":false}],"gated":[]}},"track":{"inputs":{"tracker":"Sam31MultiplexTracker","model_id":"/mnt/ceph3/ec/jacky/sam3.1","noun_phrases":["cyclist in front","cyclist in pink jersey"],"expressions_in":2,"frames":57,"sample_fps":4.0,"frame_size":"1920x1080","timestamps":[0.0,0.25025,0.5005,0.75075,1.001,1.25125,1.5015,1.75175,2.002,2.25225,2.5025,2.75275,3.003,3.25325,3.5035,3.7537499999999997,4.004,4.25425,4.5045,4.75475,5.005,5.25525,5.5055,5.75575,6.006,6.25625,6.5065,6.756749999999999,7.007,7.25725,7.507499999999999,7.75775,8.008,8.25825,8.5085,8.75875,9.009,9.25925,9.5095,9.75975,10.01,10.26025,10.5105,10.76075,11.011,11.26125,11.5115,11.76175,12.012,12.26225,12.5125,12.762749999999999,13.013,13.26325,13.513499999999999,13.76375,14.014],"session":"one session over 57 JPEG frames; then per phrase: reset_session, add_prompt(text) on frame 0, propagate_in_video","passes":2},"outputs":{"raw_tracks":[{"noun_phrase":"cyclist in front","object_id":0,"observations":[{"timestamp":0.0,"box":[0.4385416805744171,0.412962943315506,0.4802083484828472,0.6861110925674438],"occluded":false},{"timestamp":0.25025,"box":[0.4385416805744171,0.4046296179294586,0.4843750149011612,0.6851851642131805],"occluded":false},{"timestamp":0.5005,"box":[0.4270833432674408,0.4000000059604645,0.4729166775941849,0.6842592656612396],"occluded":false},{"timestamp":0.75075,"box":[0.41822919249534607,0.4055555462837219,0.47708336263895035,0.6712962687015533],"occluded":false},{"timestamp":1.001,"box":[0.41041669249534607,0.4037036895751953,0.46406252682209015,0.6925925612449646],"occluded":false},{"timestamp":1.25125,"box":[0.40260419249534607,0.4037036895751953,0.46927086263895035,0.6925925612449646],"occluded":false},{"timestamp":1.5015,"box":[0.3994791805744171,0.4009259045124054,0.47031252086162567,0.6851851642131805],"occluded":false},{"timestamp":1.75175,"box":[0.39010417461395264,0.4037036895751953,0.4505208432674408,0.6611110866069794],"occluded":false},{"timestamp":2.002,"box":[0.39010417461395264,0.40740740299224854,0.45625001192092896,0.6527777761220932],"occluded":false},{"timestamp":2.25225,"box":[0.3958333432674408,0.41111111640930176,0.46041668206453323,0.6842592656612396],"occluded":false},{"timestamp":2.5025,"box":[0.40416669845581055,0.4175925850868225,0.4567708671092987,0.6648148000240326],"occluded":false},{"timestamp":2.75275,"box":[0.4098958671092987,0.4175925850868225,0.460416704416275,0.6574074029922485],"occluded":false},{"timestamp":3.003,"box":[0.40937501192092896,0.42222222685813904,0.4651041813194752,0.6740740835666656],"occluded":false},{"timestamp":3.25325,"box":[0.4203125238418579,0.42407405376434326,0.47187502682209015,0.6685184985399246],"occluded":false},{"timestamp":3.5035,"box":[0.42760419845581055,0.42407405376434326,0.4812500327825546,0.6527777463197708],"occluded":false},{"timestamp":3.7537499999999997,"box":[0.42500001192092896,0.4268518388271332,0.47708334773778915,0.6685184985399246],"occluded":false},{"timestamp":4.004,"box":[0.43802085518836975,0.4259259104728699,0.4869791902601719,0.6675925701856613],"occluded":false},{"timestamp":4.25425,"box":[0.44166669249534607,0.4259259104728699,0.49010419473052025,0.6509259045124054],"occluded":false},{"timestamp":4.5045,"box":[0.4375000298023224,0.4277777671813965,0.4614583645015955,0.6259259134531021],"occluded":false},{"timestamp":4.75475,"box":[0.4442708492279053,0.4416666626930237,0.47500001825392246,0.6703703552484512],"occluded":false},{"timestamp":5.005,"box":[0.45104169845581055,0.460185170173645,0.47708336636424065,0.6398147940635681],"occluded":false},{"timestamp":5.25525,"box":[0.452604204416275,0.4166666567325592,0.4796875398606062,0.6416666507720947],"occluded":false},{"timestamp":5.5055,"box":[0.45468753576278687,0.41018518805503845,0.482291704043746,0.6805555522441864],"occluded":false},{"timestamp":5.75575,"box":[0.45625001192092896,0.40648147463798523,0.5088541805744171,0.6787036955356598],"occluded":false},{"timestamp":6.006,"box":[0.46250003576278687,0.39814814925193787,0.602083370089531,0.6638888716697693],"occluded":false},{"timestamp":6.25625,"box":[0.46666669845581055,0.39444443583488464,0.602083370089531,0.6749999821186066],"occluded":false},{"timestamp":6.5065,"box":[0.4713541865348816,0.39537036418914795,0.6161458641290665,0.6777777671813965],"occluded":false},{"timestamp":6.756749999999999,"box":[0.47343751788139343,0.39351850748062134,0.6437500268220901,0.6851851642131805],"occluded":false},{"timestamp":7.007,"box":[0.46302086114883423,0.39259257912635803,0.6437500417232513,0.6833333075046539],"occluded":false},{"timestamp":7.25725,"box":[0.46145835518836975,0.3916666507720947,0.5192708596587181,0.6638888716697693],"occluded":false},{"timestamp":7.507499999999999,"box":[0.4520833492279053,0.3907407224178314,0.5088541842997074,0.6898148059844971],"occluded":false},{"timestamp":7.75775,"box":[0.44062501192092896,0.44814813137054443,0.4723958484828472,0.6361110955476761],"occluded":false},{"timestamp":8.008,"box":[0.4333333671092987,0.46759259700775146,0.4630208685994148,0.6833333373069763],"occluded":false},{"timestamp":8.25825,"box":[0.44739586114883423,0.550000011920929,0.4614583617076278,0.6138888970017433],"occluded":false},{"timestamp":8.5085,"box":[0.40052086114883423,0.5435184836387634,0.43385419622063637,0.623148113489151],"occluded":false},{"timestamp":9.009,"box":[0.39375001192092896,0.47685185074806213,0.408854179084301,0.5722222179174423],"occluded":false},{"timestamp":9.75975,"box":[0.4593750238418579,0.4842592477798462,0.47864585742354393,0.6648148000240326],"occluded":false},{"timestamp":10.01,"box":[0.46302086114883423,0.44907405972480774,0.4875000286847353,0.6694444268941879],"occluded":false},{"timestamp":10.26025,"box":[0.444791704416275,0.4453703761100769,0.47968753799796104,0.6379629671573639],"occluded":false},{"timestamp":10.5105,"box":[0.44947919249534607,0.4555555582046509,0.4854166954755783,0.688888892531395],"occluded":false},{"timestamp":10.76075,"box":[0.476041704416275,0.45092591643333435,0.5020833723247051,0.6416666507720947],"occluded":false},{"timestamp":11.011,"box":[0.47812503576278687,0.4435185194015503,0.508333370089531,0.6509259194135666],"occluded":false},{"timestamp":11.26125,"box":[0.46666669845581055,0.43703702092170715,0.5145833678543568,0.6712962687015533],"occluded":false},{"timestamp":11.5115,"box":[0.4723958671092987,0.4435185194015503,0.518750037997961,0.6805555522441864],"occluded":false},{"timestamp":11.76175,"box":[0.4723958671092987,0.4018518328666687,0.518750037997961,0.6740740537643433],"occluded":false},{"timestamp":12.012,"box":[0.4755208492279053,0.442592591047287,0.5213541835546494,0.6546296179294586],"occluded":false},{"timestamp":12.26225,"box":[0.460416704416275,0.4555555582046509,0.5083333738148212,0.6620370298624039],"occluded":false},{"timestamp":12.5125,"box":[0.48125001788139343,0.5314814448356628,0.49843751825392246,0.6694444119930267],"occluded":false},{"timestamp":12.762749999999999,"box":[0.48489585518836975,0.42407405376434326,0.5223958566784859,0.5546296089887619],"occluded":false},{"timestamp":13.013,"box":[0.47187501192092896,0.4148148000240326,0.49687501415610313,0.6444444209337234],"occluded":false},{"timestamp":13.26325,"box":[0.48906251788139343,0.44999998807907104,0.5093750189989805,0.6546296179294586],"occluded":false},{"timestamp":13.513499999999999,"box":[0.483854204416275,0.44907405972480774,0.5015625394880772,0.6138888746500015],"occluded":false},{"timestamp":13.76375,"box":[0.47968751192092896,0.42314815521240234,0.5052083469927311,0.6574074029922485],"occluded":false},{"timestamp":14.014,"box":[0.47291669249534607,0.42314815521240234,0.49791669473052025,0.631481483578682],"occluded":false}],"scores":[0.8709677457809448,0.8709677457809448,0.8709677457809448,0.8709677457809448,0.8709677457809448,0.8709677457809448,0.8709677457809448,0.8709677457809448,0.8709677457809448,0.8709677457809448,0.8709677457809448,0.8709677457809448,0.8709677457809448,0.8709677457809448,0.8709677457809448,0.8709677457809448,0.8709677457809448,0.8709677457809448,0.8709677457809448,0.8709677457809448,0.8709677457809448,0.8709677457809448,0.8709677457809448,0.8709677457809448,0.8709677457809448,0.8709677457809448,0.8709677457809448,0.8709677457809448,0.8709677457809448,0.8709677457809448,0.8709677457809448,0.8709677457809448,0.8709677457809448,0.8709677457809448,0.8709677457809448,0.8709677457809448,0.8709677457809448,0.8709677457809448,0.8709677457809448,0.8709677457809448,0.8709677457809448,0.8709677457809448,0.8709677457809448,0.8709677457809448,0.8709677457809448,0.8709677457809448,0.8709677457809448,0.8709677457809448,0.8709677457809448,0.8709677457809448,0.8709677457809448,0.8709677457809448,0.8709677457809448,0.8709677457809448]},{"noun_phrase":"cyclist in front","object_id":1,"observations":[{"timestamp":0.0,"box":[0.5869792103767395,0.3722222149372101,0.6781250461935997,0.6888888776302338],"occluded":false},{"timestamp":0.25025,"box":[0.5979167222976685,0.36574074625968933,0.6802083924412727,0.6833333373069763],"occluded":false},{"timestamp":0.5005,"box":[0.5958333611488342,0.37037035822868347,0.6786458641290665,0.6666666567325592],"occluded":false},{"timestamp":0.75075,"box":[0.5963541865348816,0.3759259283542633,0.6750000268220901,0.6583333313465118],"occluded":false},{"timestamp":1.001,"box":[0.5932291746139526,0.3768518567085266,0.6687500104308128,0.6675925850868225],"occluded":false},{"timestamp":1.25125,"box":[0.5864583849906921,0.385185182094574,0.6588542237877846,0.5537036955356598],"occluded":false},{"timestamp":1.5015,"box":[0.5843750238418579,0.39351850748062134,0.65416669100523,0.6481481194496155],"occluded":false},{"timestamp":1.75175,"box":[0.5671875476837158,0.39537036418914795,0.634895883500576,0.6648148000240326],"occluded":false},{"timestamp":2.002,"box":[0.5645833611488342,0.39537036418914795,0.6307291984558105,0.6712962985038757],"occluded":false},{"timestamp":2.25225,"box":[0.5630208849906921,0.39907407760620117,0.6260417252779007,0.6518518626689911],"occluded":false},{"timestamp":2.5025,"box":[0.5593750476837158,0.4037036895751953,0.6223958879709244,0.6499999761581421],"occluded":false},{"timestamp":2.75275,"box":[0.5598958730697632,0.4037036895751953,0.6171875409781933,0.6601851582527161],"occluded":false},{"timestamp":3.003,"box":[0.5588542222976685,0.4055555462837219,0.6098958924412727,0.6499999910593033],"occluded":false},{"timestamp":3.25325,"box":[0.5510417222976685,0.412962943315506,0.6072917245328426,0.6398147940635681],"occluded":false},{"timestamp":3.5035,"box":[0.565625011920929,0.4444444477558136,0.603645846247673,0.6129629611968994],"occluded":false},{"timestamp":3.7537499999999997,"box":[0.5640625357627869,0.5435184836387634,0.5750000365078449,0.6527777388691902],"occluded":false},{"timestamp":4.004,"box":[0.5671875476837158,0.47592592239379883,0.5817708820104599,0.5879629552364349],"occluded":false},{"timestamp":4.25425,"box":[0.5593750476837158,0.4731481373310089,0.584895882755518,0.6249999850988388],"occluded":false},{"timestamp":4.5045,"box":[0.5463541746139526,0.49166664481163025,0.5864583440124989,0.6509259045124054],"occluded":false},{"timestamp":4.75475,"box":[0.5385416746139526,0.44999998807907104,0.5864583440124989,0.6518518328666687],"occluded":false},{"timestamp":5.005,"box":[0.5432292222976685,0.4314814805984497,0.5864583924412727,0.635185182094574],"occluded":false},{"timestamp":5.25525,"box":[0.5489583611488342,0.4277777671813965,0.5697916951030493,0.5703703612089157],"occluded":false},{"timestamp":5.5055,"box":[0.5604166984558105,0.42499998211860657,0.5781250335276127,0.5814814567565918],"occluded":false},{"timestamp":5.75575,"box":[0.5770833492279053,0.45092591643333435,0.5968750175088644,0.5305555462837219],"occluded":false},{"timestamp":6.25625,"box":[0.5901042222976685,0.5037037134170532,0.6052083894610405,0.5407407507300377],"occluded":false},{"timestamp":7.25725,"box":[0.6015625596046448,0.43888887763023376,0.6260417271405458,0.527777761220932],"occluded":false},{"timestamp":7.507499999999999,"box":[0.5843750238418579,0.4166666567325592,0.60729169100523,0.5268518403172493],"occluded":false},{"timestamp":7.75775,"box":[0.5671875476837158,0.4185185134410858,0.59218754991889,0.6444444358348846],"occluded":false},{"timestamp":8.008,"box":[0.5552083849906921,0.4194444417953491,0.5750000532716513,0.6240740716457367],"occluded":false},{"timestamp":8.25825,"box":[0.5416666865348816,0.42129629850387573,0.5921875238418579,0.6546296328306198],"occluded":false},{"timestamp":8.75875,"box":[0.5244792103767395,0.4296296238899231,0.5718750469386578,0.635185182094574],"occluded":false},{"timestamp":9.009,"box":[0.5062500238418579,0.4259259104728699,0.55520835891366,0.6518518328666687],"occluded":false},{"timestamp":9.25925,"box":[0.31041666865348816,0.4268518388271332,0.5572916865348816,0.6537036895751953],"occluded":false},{"timestamp":9.5095,"box":[0.503125011920929,0.4296296238899231,0.5536458492279053,0.635185182094574],"occluded":false},{"timestamp":9.75975,"box":[0.46302086114883423,0.4305555522441864,0.5484375357627869,0.6398148089647293],"occluded":false},{"timestamp":10.01,"box":[0.5010417103767395,0.4305555522441864,0.5489583797752857,0.64907406270504],"occluded":false},{"timestamp":10.26025,"box":[0.4843750298023224,0.43703702092170715,0.5317708663642406,0.6416666507720947],"occluded":false},{"timestamp":10.5105,"box":[0.4843750298023224,0.43981480598449707,0.5307292006909847,0.6342592537403107],"occluded":false},{"timestamp":10.76075,"box":[0.5005208849906921,0.43703702092170715,0.5401042215526104,0.6407407224178314],"occluded":false},{"timestamp":11.011,"box":[0.5036458373069763,0.43981480598449707,0.5401041731238365,0.6379629522562027],"occluded":false},{"timestamp":11.26125,"box":[0.5041667222976685,0.43981480598449707,0.5395833924412727,0.6212962865829468],"occluded":false},{"timestamp":11.5115,"box":[0.5145833492279053,0.4444444477558136,0.5375000163912773,0.635185182094574],"occluded":false},{"timestamp":11.76175,"box":[0.5135416984558105,0.4462962746620178,0.5401041992008686,0.5981481224298477],"occluded":false},{"timestamp":12.012,"box":[0.5119792222976685,0.44814813137054443,0.5406250562518835,0.6222222000360489],"occluded":false},{"timestamp":12.26225,"box":[0.5026041865348816,0.45092591643333435,0.5364583544433117,0.6222222149372101],"occluded":false},{"timestamp":12.5125,"box":[0.499479204416275,0.44814813137054443,0.5333333723247051,0.6277777552604675],"occluded":false},{"timestamp":12.762749999999999,"box":[0.4906250238418579,0.4277777671813965,0.5229166932404041,0.6185185015201569],"occluded":false},{"timestamp":13.013,"box":[0.4911458492279053,0.45277777314186096,0.5208333507180214,0.6166666597127914],"occluded":false},{"timestamp":13.26325,"box":[0.5,0.45370370149612427,0.5296875014901161,0.6240740716457367],"occluded":false},{"timestamp":13.513499999999999,"box":[0.4989583492279053,0.4546296298503876,0.5260416846722364,0.6175925880670547],"occluded":false},{"timestamp":13.76375,"box":[0.49635419249534607,0.46296295523643494,0.5244791936129332,0.6064814776182175],"occluded":false},{"timestamp":14.014,"box":[0.49010419845581055,0.460185170173645,0.5151042006909847,0.6166666448116302],"occluded":false}],"scores":[0.9064327478408813,0.9064327478408813,0.9064327478408813,0.9064327478408813,0.9064327478408813,0.9064327478408813,0.9064327478408813,0.9064327478408813,0.9064327478408813,0.9064327478408813,0.9064327478408813,0.9064327478408813,0.9064327478408813,0.9064327478408813,0.9064327478408813,0.9064327478408813,0.9064327478408813,0.9064327478408813,0.9064327478408813,0.9064327478408813,0.9064327478408813,0.9064327478408813,0.9064327478408813,0.9064327478408813,0.9064327478408813,0.9064327478408813,0.9064327478408813,0.9064327478408813,0.9064327478408813,0.9064327478408813,0.9064327478408813,0.9064327478408813,0.9064327478408813,0.9064327478408813,0.9064327478408813,0.9064327478408813,0.9064327478408813,0.9064327478408813,0.9064327478408813,0.9064327478408813,0.9064327478408813,0.9064327478408813,0.9064327478408813,0.9064327478408813,0.9064327478408813,0.9064327478408813,0.9064327478408813,0.9064327478408813,0.9064327478408813,0.9064327478408813,0.9064327478408813,0.9064327478408813]},{"noun_phrase":"cyclist in front","object_id":2,"observations":[{"timestamp":0.0,"box":[0.4802083671092987,0.2092592567205429,0.6203125417232513,0.838888868689537],"occluded":false},{"timestamp":0.25025,"box":[0.4802083671092987,0.206481471657753,0.6171875447034836,0.8064814358949661],"occluded":false},{"timestamp":0.5005,"box":[0.46927085518836975,0.2101851850748062,0.6031250357627869,0.7935184985399246],"occluded":false},{"timestamp":0.75075,"box":[0.46145835518836975,0.2203703671693802,0.5921875238418579,0.7879629284143448],"occluded":false},{"timestamp":1.001,"box":[0.45104169845581055,0.23333333432674408,0.5848958790302277,0.8129629343748093],"occluded":false},{"timestamp":1.25125,"box":[0.44322919845581055,0.2342592477798462,0.5677083730697632,0.7731481194496155],"occluded":false},{"timestamp":1.5015,"box":[0.4437500238418579,0.2370370328426361,0.5744791924953461,0.8055555522441864],"occluded":false},{"timestamp":1.75175,"box":[0.4255208671092987,0.23888888955116272,0.555208370089531,0.8166666328907013],"occluded":false},{"timestamp":2.002,"box":[0.4229166805744171,0.24166665971279144,0.5614583492279053,0.7861111015081406],"occluded":false},{"timestamp":2.25225,"box":[0.42500001192092896,0.25,0.5635416805744171,0.7749999761581421],"occluded":false},{"timestamp":2.5025,"box":[0.43020835518836975,0.2537037134170532,0.5635416954755783,0.8046296238899231],"occluded":false},{"timestamp":2.75275,"box":[0.4442708492279053,0.2888888716697693,0.5708333551883698,0.7981480956077576],"occluded":false},{"timestamp":3.003,"box":[0.44843751192092896,0.28703704476356506,0.5723958536982536,0.7583333253860474],"occluded":false},{"timestamp":3.25325,"box":[0.4661458432674408,0.25740739703178406,0.5927083492279053,0.7722221910953522],"occluded":false},{"timestamp":3.5035,"box":[0.4791666865348816,0.2620370388031006,0.6072916984558105,0.7935184836387634],"occluded":false},{"timestamp":3.7537499999999997,"box":[0.4739583432674408,0.2638888955116272,0.6015625149011612,0.7833333015441895],"occluded":false},{"timestamp":4.004,"box":[0.48125001788139343,0.2527777850627899,0.6093750298023224,0.7537036836147308],"occluded":false},{"timestamp":4.25425,"box":[0.47083336114883423,0.2509259283542633,0.6046875417232513,0.7888889014720917],"occluded":false},{"timestamp":4.5045,"box":[0.44166669249534607,0.24444444477558136,0.5812500268220901,0.8138888627290726],"occluded":false},{"timestamp":4.75475,"box":[0.44166669249534607,0.23796296119689941,0.5833333730697632,0.7972221970558167],"occluded":false},{"timestamp":5.005,"box":[0.4479166865348816,0.22870369255542755,0.5921875238418579,0.781481459736824],"occluded":false},{"timestamp":5.25525,"box":[0.4520833492279053,0.2222222238779068,0.6015625298023224,0.8138888627290726],"occluded":false},{"timestamp":5.5055,"box":[0.4687500298023224,0.2231481373310089,0.617708370089531,0.8249999582767487],"occluded":false},{"timestamp":5.75575,"box":[0.49427086114883423,0.21666666865348816,0.6421875357627869,0.7861110866069794],"occluded":false},{"timestamp":6.006,"box":[0.5250000357627869,0.206481471657753,0.6781250387430191,0.8074073940515518],"occluded":false},{"timestamp":6.25625,"box":[0.5473958849906921,0.20462962985038757,0.7062500566244125,0.8212963044643402],"occluded":false},{"timestamp":6.5065,"box":[0.5546875,0.2101851850748062,0.7156250029802322,0.7824074178934097],"occluded":false},{"timestamp":6.756749999999999,"box":[0.5463541746139526,0.20277777314186096,0.7156250178813934,0.7824073731899261],"occluded":false},{"timestamp":7.007,"box":[0.46302086114883423,0.2074074000120163,0.7005208730697632,0.8268518298864365],"occluded":false},{"timestamp":7.25725,"box":[0.5140625238418579,0.21203702688217163,0.6697916984558105,0.8231481313705444],"occluded":false},{"timestamp":7.507499999999999,"box":[0.48645836114883423,0.19166666269302368,0.6453125327825546,0.7842592597007751],"occluded":false},{"timestamp":7.75775,"box":[0.4598958492279053,0.32314813137054443,0.6250000298023224,0.824999988079071],"occluded":false},{"timestamp":8.008,"box":[0.4385416805744171,0.32592591643333435,0.6062500178813934,0.8444444239139557],"occluded":false},{"timestamp":8.25825,"box":[0.4130208492279053,0.3583333194255829,0.5854166895151138,0.8240740597248077],"occluded":false},{"timestamp":8.5085,"box":[0.3968750238418579,0.10185185074806213,0.6364583671092987,0.8240740597248077],"occluded":false},{"timestamp":8.75875,"box":[0.3515625298023224,0.18518517911434174,0.5375000387430191,0.8527777642011642],"occluded":false},{"timestamp":9.009,"box":[0.3411458432674408,0.192592591047287,0.525520846247673,0.8398148119449615],"occluded":false},{"timestamp":9.25925,"box":[0.3296875059604645,0.1787036955356598,0.5166666805744171,0.8064814507961273],"occluded":false},{"timestamp":9.5095,"box":[0.31041666865348816,0.19074073433876038,0.49166667461395264,0.8425925672054291],"occluded":false},{"timestamp":9.75975,"box":[0.30520835518836975,0.19722221791744232,0.4854166954755783,0.8462962955236435],"occluded":false},{"timestamp":10.01,"box":[0.3036458492279053,0.20000000298023224,0.4843750298023224,0.8185185343027115],"occluded":false},{"timestamp":10.26025,"box":[0.2796875238418579,0.19722221791744232,0.4895833730697632,0.8185185045003891],"occluded":false},{"timestamp":10.5105,"box":[0.29635417461395264,0.20277777314186096,0.4609375149011612,0.8444444239139557],"occluded":false},{"timestamp":10.76075,"box":[0.31822916865348816,0.2074074000120163,0.48489584028720856,0.8435184806585312],"occluded":false},{"timestamp":11.011,"box":[0.3218750059604645,0.20185184478759766,0.4843750149011612,0.7990740537643433],"occluded":false},{"timestamp":11.26125,"box":[0.3213541805744171,0.19814814627170563,0.4843750298023224,0.8305555135011673],"occluded":false},{"timestamp":11.5115,"box":[0.30781251192092896,0.20555555820465088,0.4750000238418579,0.8499999642372131],"occluded":false},{"timestamp":11.76175,"box":[0.3098958432674408,0.19629628956317902,0.4776041805744171,0.8203703314065933],"occluded":false},{"timestamp":12.012,"box":[0.3072916865348816,0.20185184478759766,0.4765625298023224,0.8064814805984497],"occluded":false},{"timestamp":12.26225,"box":[0.32447919249534607,0.18518517911434174,0.5015625357627869,0.8398148268461227],"occluded":false},{"timestamp":12.5125,"box":[0.3229166865348816,0.18611110746860504,0.5005208551883698,0.7500000149011612],"occluded":false},{"timestamp":12.762749999999999,"box":[0.3239583373069763,0.18888889253139496,0.4932291805744171,0.7907407134771347],"occluded":false},{"timestamp":13.013,"box":[0.31666669249534607,0.1953703612089157,0.47864586114883423,0.8268518298864365],"occluded":false},{"timestamp":13.26325,"box":[0.36666667461395264,0.2074074000120163,0.5026041865348816,0.8268518298864365],"occluded":false},{"timestamp":13.513499999999999,"box":[0.33020836114883423,0.21759258210659027,0.4869792014360428,0.803703710436821],"occluded":false},{"timestamp":13.76375,"box":[0.3307291865348816,0.2203703671693802,0.4812500327825546,0.7888888865709305],"occluded":false},{"timestamp":14.014,"box":[0.3317708373069763,0.22962962090969086,0.47604167461395264,0.7999999970197678],"occluded":false}],"scores":[0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708]},{"noun_phrase":"cyclist in pink jersey","object_id":0,"observations":[{"timestamp":0.0,"box":[0.5869792103767395,0.3722222149372101,0.6781250461935997,0.6888888776302338],"occluded":false},{"timestamp":0.25025,"box":[0.5984375476837158,0.36574074625968933,0.6802083849906921,0.682407408952713],"occluded":false},{"timestamp":0.5005,"box":[0.5953125357627869,0.37037035822868347,0.6786458715796471,0.6657407283782959],"occluded":false},{"timestamp":0.75075,"box":[0.5083333849906921,0.36574074625968933,0.6750000566244125,0.6574074029922485],"occluded":false},{"timestamp":1.001,"box":[0.5937500596046448,0.3768518567085266,0.668750062584877,0.6657407283782959],"occluded":false},{"timestamp":1.25125,"box":[0.5864583849906921,0.385185182094574,0.6588542237877846,0.5537036955356598],"occluded":false},{"timestamp":1.5015,"box":[0.5843750238418579,0.39259257912635803,0.6546875238418579,0.6472221910953522],"occluded":false},{"timestamp":1.75175,"box":[0.5671875476837158,0.39537036418914795,0.634895883500576,0.6249999850988388],"occluded":false},{"timestamp":2.002,"box":[0.5645833611488342,0.39537036418914795,0.6307291984558105,0.6703703701496124],"occluded":false},{"timestamp":2.25225,"box":[0.5635417103767395,0.39907407760620117,0.6265625506639481,0.6509259343147278],"occluded":false},{"timestamp":2.5025,"box":[0.46666669845581055,0.39351850748062134,0.6223958730697632,0.6490740478038788],"occluded":false}],"scores":[0.7419355511665344,0.7419355511665344,0.7419355511665344,0.7419355511665344,0.7419355511665344,0.7419355511665344,0.7419355511665344,0.7419355511665344,0.7419355511665344,0.7419355511665344,null]}],"tracklets":[{"noun_phrase":"cyclist in front","object_id":0,"presentation_id":1,"observations":[{"timestamp":0.0,"box":[0.4385416805744171,0.412962943315506,0.4802083484828472,0.6861110925674438],"occluded":false},{"timestamp":0.25025,"box":[0.4385416805744171,0.4046296179294586,0.4843750149011612,0.6851851642131805],"occluded":false},{"timestamp":0.5005,"box":[0.4270833432674408,0.4000000059604645,0.4729166775941849,0.6842592656612396],"occluded":false},{"timestamp":0.75075,"box":[0.41822919249534607,0.4055555462837219,0.47708336263895035,0.6712962687015533],"occluded":false},{"timestamp":1.001,"box":[0.41041669249534607,0.4037036895751953,0.46406252682209015,0.6925925612449646],"occluded":false},{"timestamp":1.25125,"box":[0.40260419249534607,0.4037036895751953,0.46927086263895035,0.6925925612449646],"occluded":false},{"timestamp":1.5015,"box":[0.3994791805744171,0.4009259045124054,0.47031252086162567,0.6851851642131805],"occluded":false},{"timestamp":1.75175,"box":[0.39010417461395264,0.4037036895751953,0.4505208432674408,0.6611110866069794],"occluded":false},{"timestamp":2.002,"box":[0.39010417461395264,0.40740740299224854,0.45625001192092896,0.6527777761220932],"occluded":false},{"timestamp":2.25225,"box":[0.3958333432674408,0.41111111640930176,0.46041668206453323,0.6842592656612396],"occluded":false},{"timestamp":2.5025,"box":[0.40416669845581055,0.4175925850868225,0.4567708671092987,0.6648148000240326],"occluded":false},{"timestamp":2.75275,"box":[0.4098958671092987,0.4175925850868225,0.460416704416275,0.6574074029922485],"occluded":false},{"timestamp":3.003,"box":[0.40937501192092896,0.42222222685813904,0.4651041813194752,0.6740740835666656],"occluded":false},{"timestamp":3.25325,"box":[0.4203125238418579,0.42407405376434326,0.47187502682209015,0.6685184985399246],"occluded":false},{"timestamp":3.5035,"box":[0.42760419845581055,0.42407405376434326,0.4812500327825546,0.6527777463197708],"occluded":false},{"timestamp":3.7537499999999997,"box":[0.42500001192092896,0.4268518388271332,0.47708334773778915,0.6685184985399246],"occluded":false},{"timestamp":4.004,"box":[0.43802085518836975,0.4259259104728699,0.4869791902601719,0.6675925701856613],"occluded":false},{"timestamp":4.25425,"box":[0.44166669249534607,0.4259259104728699,0.49010419473052025,0.6509259045124054],"occluded":false},{"timestamp":4.5045,"box":[0.4375000298023224,0.4277777671813965,0.4614583645015955,0.6259259134531021],"occluded":false},{"timestamp":4.75475,"box":[0.4442708492279053,0.4416666626930237,0.47500001825392246,0.6703703552484512],"occluded":false},{"timestamp":5.005,"box":[0.45104169845581055,0.460185170173645,0.47708336636424065,0.6398147940635681],"occluded":false},{"timestamp":5.25525,"box":[0.452604204416275,0.4166666567325592,0.4796875398606062,0.6416666507720947],"occluded":false},{"timestamp":5.5055,"box":[0.45468753576278687,0.41018518805503845,0.482291704043746,0.6805555522441864],"occluded":false},{"timestamp":5.75575,"box":[0.45625001192092896,0.40648147463798523,0.5088541805744171,0.6787036955356598],"occluded":false},{"timestamp":6.006,"box":[0.46250003576278687,0.39814814925193787,0.602083370089531,0.6638888716697693],"occluded":false},{"timestamp":6.25625,"box":[0.46666669845581055,0.39444443583488464,0.602083370089531,0.6749999821186066],"occluded":false},{"timestamp":6.5065,"box":[0.4713541865348816,0.39537036418914795,0.6161458641290665,0.6777777671813965],"occluded":false},{"timestamp":6.756749999999999,"box":[0.47343751788139343,0.39351850748062134,0.6437500268220901,0.6851851642131805],"occluded":false},{"timestamp":7.007,"box":[0.46302086114883423,0.39259257912635803,0.6437500417232513,0.6833333075046539],"occluded":false},{"timestamp":7.25725,"box":[0.46145835518836975,0.3916666507720947,0.5192708596587181,0.6638888716697693],"occluded":false},{"timestamp":7.507499999999999,"box":[0.4520833492279053,0.3907407224178314,0.5088541842997074,0.6898148059844971],"occluded":false},{"timestamp":7.75775,"box":[0.44062501192092896,0.44814813137054443,0.4723958484828472,0.6361110955476761],"occluded":false},{"timestamp":8.008,"box":[0.4333333671092987,0.46759259700775146,0.4630208685994148,0.6833333373069763],"occluded":false},{"timestamp":8.25825,"box":[0.44739586114883423,0.550000011920929,0.4614583617076278,0.6138888970017433],"occluded":false},{"timestamp":8.5085,"box":[0.40052086114883423,0.5435184836387634,0.43385419622063637,0.623148113489151],"occluded":false},{"timestamp":9.009,"box":[0.39375001192092896,0.47685185074806213,0.408854179084301,0.5722222179174423],"occluded":false},{"timestamp":9.75975,"box":[0.4593750238418579,0.4842592477798462,0.47864585742354393,0.6648148000240326],"occluded":false},{"timestamp":10.01,"box":[0.46302086114883423,0.44907405972480774,0.4875000286847353,0.6694444268941879],"occluded":false},{"timestamp":10.26025,"box":[0.444791704416275,0.4453703761100769,0.47968753799796104,0.6379629671573639],"occluded":false},{"timestamp":10.5105,"box":[0.44947919249534607,0.4555555582046509,0.4854166954755783,0.688888892531395],"occluded":false},{"timestamp":10.76075,"box":[0.476041704416275,0.45092591643333435,0.5020833723247051,0.6416666507720947],"occluded":false},{"timestamp":11.011,"box":[0.47812503576278687,0.4435185194015503,0.508333370089531,0.6509259194135666],"occluded":false},{"timestamp":11.26125,"box":[0.46666669845581055,0.43703702092170715,0.5145833678543568,0.6712962687015533],"occluded":false},{"timestamp":11.5115,"box":[0.4723958671092987,0.4435185194015503,0.518750037997961,0.6805555522441864],"occluded":false},{"timestamp":11.76175,"box":[0.4723958671092987,0.4018518328666687,0.518750037997961,0.6740740537643433],"occluded":false},{"timestamp":12.012,"box":[0.4755208492279053,0.442592591047287,0.5213541835546494,0.6546296179294586],"occluded":false},{"timestamp":12.26225,"box":[0.460416704416275,0.4555555582046509,0.5083333738148212,0.6620370298624039],"occluded":false},{"timestamp":12.5125,"box":[0.48125001788139343,0.5314814448356628,0.49843751825392246,0.6694444119930267],"occluded":false},{"timestamp":12.762749999999999,"box":[0.48489585518836975,0.42407405376434326,0.5223958566784859,0.5546296089887619],"occluded":false},{"timestamp":13.013,"box":[0.47187501192092896,0.4148148000240326,0.49687501415610313,0.6444444209337234],"occluded":false},{"timestamp":13.26325,"box":[0.48906251788139343,0.44999998807907104,0.5093750189989805,0.6546296179294586],"occluded":false},{"timestamp":13.513499999999999,"box":[0.483854204416275,0.44907405972480774,0.5015625394880772,0.6138888746500015],"occluded":false},{"timestamp":13.76375,"box":[0.47968751192092896,0.42314815521240234,0.5052083469927311,0.6574074029922485],"occluded":false},{"timestamp":14.014,"box":[0.47291669249534607,0.42314815521240234,0.49791669473052025,0.631481483578682],"occluded":false}],"mean_score":0.8709677457809448},{"noun_phrase":"cyclist in front","object_id":2,"presentation_id":2,"observations":[{"timestamp":0.0,"box":[0.4802083671092987,0.2092592567205429,0.6203125417232513,0.838888868689537],"occluded":false},{"timestamp":0.25025,"box":[0.4802083671092987,0.206481471657753,0.6171875447034836,0.8064814358949661],"occluded":false},{"timestamp":0.5005,"box":[0.46927085518836975,0.2101851850748062,0.6031250357627869,0.7935184985399246],"occluded":false},{"timestamp":0.75075,"box":[0.46145835518836975,0.2203703671693802,0.5921875238418579,0.7879629284143448],"occluded":false},{"timestamp":1.001,"box":[0.45104169845581055,0.23333333432674408,0.5848958790302277,0.8129629343748093],"occluded":false},{"timestamp":1.25125,"box":[0.44322919845581055,0.2342592477798462,0.5677083730697632,0.7731481194496155],"occluded":false},{"timestamp":1.5015,"box":[0.4437500238418579,0.2370370328426361,0.5744791924953461,0.8055555522441864],"occluded":false},{"timestamp":1.75175,"box":[0.4255208671092987,0.23888888955116272,0.555208370089531,0.8166666328907013],"occluded":false},{"timestamp":2.002,"box":[0.4229166805744171,0.24166665971279144,0.5614583492279053,0.7861111015081406],"occluded":false},{"timestamp":2.25225,"box":[0.42500001192092896,0.25,0.5635416805744171,0.7749999761581421],"occluded":false},{"timestamp":2.5025,"box":[0.43020835518836975,0.2537037134170532,0.5635416954755783,0.8046296238899231],"occluded":false},{"timestamp":2.75275,"box":[0.4442708492279053,0.2888888716697693,0.5708333551883698,0.7981480956077576],"occluded":false},{"timestamp":3.003,"box":[0.44843751192092896,0.28703704476356506,0.5723958536982536,0.7583333253860474],"occluded":false},{"timestamp":3.25325,"box":[0.4661458432674408,0.25740739703178406,0.5927083492279053,0.7722221910953522],"occluded":false},{"timestamp":3.5035,"box":[0.4791666865348816,0.2620370388031006,0.6072916984558105,0.7935184836387634],"occluded":false},{"timestamp":3.7537499999999997,"box":[0.4739583432674408,0.2638888955116272,0.6015625149011612,0.7833333015441895],"occluded":false},{"timestamp":4.004,"box":[0.48125001788139343,0.2527777850627899,0.6093750298023224,0.7537036836147308],"occluded":false},{"timestamp":4.25425,"box":[0.47083336114883423,0.2509259283542633,0.6046875417232513,0.7888889014720917],"occluded":false},{"timestamp":4.5045,"box":[0.44166669249534607,0.24444444477558136,0.5812500268220901,0.8138888627290726],"occluded":false},{"timestamp":4.75475,"box":[0.44166669249534607,0.23796296119689941,0.5833333730697632,0.7972221970558167],"occluded":false},{"timestamp":5.005,"box":[0.4479166865348816,0.22870369255542755,0.5921875238418579,0.781481459736824],"occluded":false},{"timestamp":5.25525,"box":[0.4520833492279053,0.2222222238779068,0.6015625298023224,0.8138888627290726],"occluded":false},{"timestamp":5.5055,"box":[0.4687500298023224,0.2231481373310089,0.617708370089531,0.8249999582767487],"occluded":false},{"timestamp":5.75575,"box":[0.49427086114883423,0.21666666865348816,0.6421875357627869,0.7861110866069794],"occluded":false},{"timestamp":6.006,"box":[0.5250000357627869,0.206481471657753,0.6781250387430191,0.8074073940515518],"occluded":false},{"timestamp":6.25625,"box":[0.5473958849906921,0.20462962985038757,0.7062500566244125,0.8212963044643402],"occluded":false},{"timestamp":6.5065,"box":[0.5546875,0.2101851850748062,0.7156250029802322,0.7824074178934097],"occluded":false},{"timestamp":6.756749999999999,"box":[0.5463541746139526,0.20277777314186096,0.7156250178813934,0.7824073731899261],"occluded":false},{"timestamp":7.007,"box":[0.46302086114883423,0.2074074000120163,0.7005208730697632,0.8268518298864365],"occluded":false},{"timestamp":7.25725,"box":[0.5140625238418579,0.21203702688217163,0.6697916984558105,0.8231481313705444],"occluded":false},{"timestamp":7.507499999999999,"box":[0.48645836114883423,0.19166666269302368,0.6453125327825546,0.7842592597007751],"occluded":false},{"timestamp":7.75775,"box":[0.4598958492279053,0.32314813137054443,0.6250000298023224,0.824999988079071],"occluded":false},{"timestamp":8.008,"box":[0.4385416805744171,0.32592591643333435,0.6062500178813934,0.8444444239139557],"occluded":false},{"timestamp":8.25825,"box":[0.4130208492279053,0.3583333194255829,0.5854166895151138,0.8240740597248077],"occluded":false},{"timestamp":8.5085,"box":[0.3968750238418579,0.10185185074806213,0.6364583671092987,0.8240740597248077],"occluded":false},{"timestamp":8.75875,"box":[0.3515625298023224,0.18518517911434174,0.5375000387430191,0.8527777642011642],"occluded":false},{"timestamp":9.009,"box":[0.3411458432674408,0.192592591047287,0.525520846247673,0.8398148119449615],"occluded":false},{"timestamp":9.25925,"box":[0.3296875059604645,0.1787036955356598,0.5166666805744171,0.8064814507961273],"occluded":false},{"timestamp":9.5095,"box":[0.31041666865348816,0.19074073433876038,0.49166667461395264,0.8425925672054291],"occluded":false},{"timestamp":9.75975,"box":[0.30520835518836975,0.19722221791744232,0.4854166954755783,0.8462962955236435],"occluded":false},{"timestamp":10.01,"box":[0.3036458492279053,0.20000000298023224,0.4843750298023224,0.8185185343027115],"occluded":false},{"timestamp":10.26025,"box":[0.2796875238418579,0.19722221791744232,0.4895833730697632,0.8185185045003891],"occluded":false},{"timestamp":10.5105,"box":[0.29635417461395264,0.20277777314186096,0.4609375149011612,0.8444444239139557],"occluded":false},{"timestamp":10.76075,"box":[0.31822916865348816,0.2074074000120163,0.48489584028720856,0.8435184806585312],"occluded":false},{"timestamp":11.011,"box":[0.3218750059604645,0.20185184478759766,0.4843750149011612,0.7990740537643433],"occluded":false},{"timestamp":11.26125,"box":[0.3213541805744171,0.19814814627170563,0.4843750298023224,0.8305555135011673],"occluded":false},{"timestamp":11.5115,"box":[0.30781251192092896,0.20555555820465088,0.4750000238418579,0.8499999642372131],"occluded":false},{"timestamp":11.76175,"box":[0.3098958432674408,0.19629628956317902,0.4776041805744171,0.8203703314065933],"occluded":false},{"timestamp":12.012,"box":[0.3072916865348816,0.20185184478759766,0.4765625298023224,0.8064814805984497],"occluded":false},{"timestamp":12.26225,"box":[0.32447919249534607,0.18518517911434174,0.5015625357627869,0.8398148268461227],"occluded":false},{"timestamp":12.5125,"box":[0.3229166865348816,0.18611110746860504,0.5005208551883698,0.7500000149011612],"occluded":false},{"timestamp":12.762749999999999,"box":[0.3239583373069763,0.18888889253139496,0.4932291805744171,0.7907407134771347],"occluded":false},{"timestamp":13.013,"box":[0.31666669249534607,0.1953703612089157,0.47864586114883423,0.8268518298864365],"occluded":false},{"timestamp":13.26325,"box":[0.36666667461395264,0.2074074000120163,0.5026041865348816,0.8268518298864365],"occluded":false},{"timestamp":13.513499999999999,"box":[0.33020836114883423,0.21759258210659027,0.4869792014360428,0.803703710436821],"occluded":false},{"timestamp":13.76375,"box":[0.3307291865348816,0.2203703671693802,0.4812500327825546,0.7888888865709305],"occluded":false},{"timestamp":14.014,"box":[0.3317708373069763,0.22962962090969086,0.47604167461395264,0.7999999970197678],"occluded":false}],"mean_score":0.941605806350708},{"noun_phrase":"cyclist in front","object_id":1,"presentation_id":3,"observations":[{"timestamp":0.0,"box":[0.5869792103767395,0.3722222149372101,0.6781250461935997,0.6888888776302338],"occluded":false},{"timestamp":0.25025,"box":[0.5979167222976685,0.36574074625968933,0.6802083924412727,0.6833333373069763],"occluded":false},{"timestamp":0.5005,"box":[0.5958333611488342,0.37037035822868347,0.6786458641290665,0.6666666567325592],"occluded":false},{"timestamp":0.75075,"box":[0.5963541865348816,0.3759259283542633,0.6750000268220901,0.6583333313465118],"occluded":false},{"timestamp":1.001,"box":[0.5932291746139526,0.3768518567085266,0.6687500104308128,0.6675925850868225],"occluded":false},{"timestamp":1.25125,"box":[0.5864583849906921,0.385185182094574,0.6588542237877846,0.5537036955356598],"occluded":false},{"timestamp":1.5015,"box":[0.5843750238418579,0.39351850748062134,0.65416669100523,0.6481481194496155],"occluded":false},{"timestamp":1.75175,"box":[0.5671875476837158,0.39537036418914795,0.634895883500576,0.6648148000240326],"occluded":false},{"timestamp":2.002,"box":[0.5645833611488342,0.39537036418914795,0.6307291984558105,0.6712962985038757],"occluded":false},{"timestamp":2.25225,"box":[0.5630208849906921,0.39907407760620117,0.6260417252779007,0.6518518626689911],"occluded":false},{"timestamp":2.5025,"box":[0.5593750476837158,0.4037036895751953,0.6223958879709244,0.6499999761581421],"occluded":false},{"timestamp":2.75275,"box":[0.5598958730697632,0.4037036895751953,0.6171875409781933,0.6601851582527161],"occluded":false},{"timestamp":3.003,"box":[0.5588542222976685,0.4055555462837219,0.6098958924412727,0.6499999910593033],"occluded":false},{"timestamp":3.25325,"box":[0.5510417222976685,0.412962943315506,0.6072917245328426,0.6398147940635681],"occluded":false},{"timestamp":3.5035,"box":[0.565625011920929,0.4444444477558136,0.603645846247673,0.6129629611968994],"occluded":false},{"timestamp":3.7537499999999997,"box":[0.5640625357627869,0.5435184836387634,0.5750000365078449,0.6527777388691902],"occluded":false},{"timestamp":4.004,"box":[0.5671875476837158,0.47592592239379883,0.5817708820104599,0.5879629552364349],"occluded":false},{"timestamp":4.25425,"box":[0.5593750476837158,0.4731481373310089,0.584895882755518,0.6249999850988388],"occluded":false},{"timestamp":4.5045,"box":[0.5463541746139526,0.49166664481163025,0.5864583440124989,0.6509259045124054],"occluded":false},{"timestamp":4.75475,"box":[0.5385416746139526,0.44999998807907104,0.5864583440124989,0.6518518328666687],"occluded":false},{"timestamp":5.005,"box":[0.5432292222976685,0.4314814805984497,0.5864583924412727,0.635185182094574],"occluded":false},{"timestamp":5.25525,"box":[0.5489583611488342,0.4277777671813965,0.5697916951030493,0.5703703612089157],"occluded":false},{"timestamp":5.5055,"box":[0.5604166984558105,0.42499998211860657,0.5781250335276127,0.5814814567565918],"occluded":false},{"timestamp":5.75575,"box":[0.5770833492279053,0.45092591643333435,0.5968750175088644,0.5305555462837219],"occluded":false},{"timestamp":6.25625,"box":[0.5901042222976685,0.5037037134170532,0.6052083894610405,0.5407407507300377],"occluded":false},{"timestamp":7.25725,"box":[0.6015625596046448,0.43888887763023376,0.6260417271405458,0.527777761220932],"occluded":false},{"timestamp":7.507499999999999,"box":[0.5843750238418579,0.4166666567325592,0.60729169100523,0.5268518403172493],"occluded":false},{"timestamp":7.75775,"box":[0.5671875476837158,0.4185185134410858,0.59218754991889,0.6444444358348846],"occluded":false},{"timestamp":8.008,"box":[0.5552083849906921,0.4194444417953491,0.5750000532716513,0.6240740716457367],"occluded":false},{"timestamp":8.25825,"box":[0.5416666865348816,0.42129629850387573,0.5921875238418579,0.6546296328306198],"occluded":false},{"timestamp":8.75875,"box":[0.5244792103767395,0.4296296238899231,0.5718750469386578,0.635185182094574],"occluded":false},{"timestamp":9.009,"box":[0.5062500238418579,0.4259259104728699,0.55520835891366,0.6518518328666687],"occluded":false},{"timestamp":9.25925,"box":[0.31041666865348816,0.4268518388271332,0.5572916865348816,0.6537036895751953],"occluded":false},{"timestamp":9.5095,"box":[0.503125011920929,0.4296296238899231,0.5536458492279053,0.635185182094574],"occluded":false},{"timestamp":9.75975,"box":[0.46302086114883423,0.4305555522441864,0.5484375357627869,0.6398148089647293],"occluded":false},{"timestamp":10.01,"box":[0.5010417103767395,0.4305555522441864,0.5489583797752857,0.64907406270504],"occluded":false},{"timestamp":10.26025,"box":[0.4843750298023224,0.43703702092170715,0.5317708663642406,0.6416666507720947],"occluded":false},{"timestamp":10.5105,"box":[0.4843750298023224,0.43981480598449707,0.5307292006909847,0.6342592537403107],"occluded":false},{"timestamp":10.76075,"box":[0.5005208849906921,0.43703702092170715,0.5401042215526104,0.6407407224178314],"occluded":false},{"timestamp":11.011,"box":[0.5036458373069763,0.43981480598449707,0.5401041731238365,0.6379629522562027],"occluded":false},{"timestamp":11.26125,"box":[0.5041667222976685,0.43981480598449707,0.5395833924412727,0.6212962865829468],"occluded":false},{"timestamp":11.5115,"box":[0.5145833492279053,0.4444444477558136,0.5375000163912773,0.635185182094574],"occluded":false},{"timestamp":11.76175,"box":[0.5135416984558105,0.4462962746620178,0.5401041992008686,0.5981481224298477],"occluded":false},{"timestamp":12.012,"box":[0.5119792222976685,0.44814813137054443,0.5406250562518835,0.6222222000360489],"occluded":false},{"timestamp":12.26225,"box":[0.5026041865348816,0.45092591643333435,0.5364583544433117,0.6222222149372101],"occluded":false},{"timestamp":12.5125,"box":[0.499479204416275,0.44814813137054443,0.5333333723247051,0.6277777552604675],"occluded":false},{"timestamp":12.762749999999999,"box":[0.4906250238418579,0.4277777671813965,0.5229166932404041,0.6185185015201569],"occluded":false},{"timestamp":13.013,"box":[0.4911458492279053,0.45277777314186096,0.5208333507180214,0.6166666597127914],"occluded":false},{"timestamp":13.26325,"box":[0.5,0.45370370149612427,0.5296875014901161,0.6240740716457367],"occluded":false},{"timestamp":13.513499999999999,"box":[0.4989583492279053,0.4546296298503876,0.5260416846722364,0.6175925880670547],"occluded":false},{"timestamp":13.76375,"box":[0.49635419249534607,0.46296295523643494,0.5244791936129332,0.6064814776182175],"occluded":false},{"timestamp":14.014,"box":[0.49010419845581055,0.460185170173645,0.5151042006909847,0.6166666448116302],"occluded":false}],"mean_score":0.9064327478408813}],"dropped":{"merged_duplicate":1},"fates":[{"noun_phrase":"cyclist in front","object_id":0,"frames":54,"mean_score":0.8709677457809448,"mean_area":0.012779751802799624,"outcome":"kept","presentation_id":1,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"cyclist in front","object_id":1,"frames":52,"mean_score":0.9064327478408813,"mean_area":0.010447559170193872,"outcome":"kept","presentation_id":3,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"cyclist in front","object_id":2,"frames":57,"mean_score":0.941605806350708,"mean_area":0.09238976965920384,"outcome":"kept","presentation_id":2,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"cyclist in pink jersey","object_id":0,"frames":11,"mean_score":0.7419355511665344,"mean_area":0.024469083550309804,"outcome":"merged_duplicate","presentation_id":null,"merged_into_phrase":"cyclist in front","merged_into_object_id":1,"merge_iou":0.8750229297682336}],"gates":{"min_track_frames":3,"min_mean_score":0.5,"merge_iou":0.8,"max_tracklets":12}}},"render":{"inputs":{"tracklets":3,"as_video":true,"fps":1.0,"max_frames":30},"outputs":{"media":"/home/zzk/data/video_bbox/vlm_gemini3.5_lite/renders/--Dx9HVcQWs_chunk_001_scene_034_choices.mp4","media_bytes":4807077,"frames":[{"index":1,"timestamp":0.0,"file":"--Dx9HVcQWs_chunk_001_scene_034_choices.mp4"},{"index":2,"timestamp":1.001,"file":"--Dx9HVcQWs_chunk_001_scene_034_choices.mp4"},{"index":3,"timestamp":2.002,"file":"--Dx9HVcQWs_chunk_001_scene_034_choices.mp4"},{"index":4,"timestamp":3.003,"file":"--Dx9HVcQWs_chunk_001_scene_034_choices.mp4"},{"index":5,"timestamp":4.004,"file":"--Dx9HVcQWs_chunk_001_scene_034_choices.mp4"},{"index":6,"timestamp":5.005,"file":"--Dx9HVcQWs_chunk_001_scene_034_choices.mp4"},{"index":7,"timestamp":6.006,"file":"--Dx9HVcQWs_chunk_001_scene_034_choices.mp4"},{"index":8,"timestamp":7.007,"file":"--Dx9HVcQWs_chunk_001_scene_034_choices.mp4"},{"index":9,"timestamp":8.008,"file":"--Dx9HVcQWs_chunk_001_scene_034_choices.mp4"},{"index":10,"timestamp":9.009,"file":"--Dx9HVcQWs_chunk_001_scene_034_choices.mp4"},{"index":11,"timestamp":10.01,"file":"--Dx9HVcQWs_chunk_001_scene_034_choices.mp4"},{"index":12,"timestamp":11.011,"file":"--Dx9HVcQWs_chunk_001_scene_034_choices.mp4"},{"index":13,"timestamp":12.012,"file":"--Dx9HVcQWs_chunk_001_scene_034_choices.mp4"},{"index":14,"timestamp":13.013,"file":"--Dx9HVcQWs_chunk_001_scene_034_choices.mp4"},{"index":15,"timestamp":14.014,"file":"--Dx9HVcQWs_chunk_001_scene_034_choices.mp4"}]}},"select":{"inputs":{"model":"gemini-3.5-flash-lite","temperature":0.0,"as_video":true,"media_bytes":4807077,"media":["--Dx9HVcQWs_chunk_001_scene_034_choices.mp4"],"choice_frames":15,"system_prompt":"You are labelling video data for referring multi-object tracking.\n\nYou will see a numbered sequence of frames from one clip, in time order. Each frame has its own number and timestamp printed in the top-left corner. Candidate objects have been detected for you and drawn on the frames as coloured boxes, each labelled with a number. The same object keeps the same number and the same colour on every frame. Below each frame is a strip of enlarged crops of those same numbered boxes, so you can identify a small or distant object.\n\nFor each description you are given, answer two questions:\n\n1. MATCHING_IDS — which of the numbered objects the description refers to. List every number that matches, or an EMPTY LIST if none of them do. An empty list is a correct and expected answer: some descriptions are deliberately about something that is not in this clip. Do not pick a number just to have picked one, and do not pick a number for an object that merely resembles the description.\n\n2. START_FRAME and END_FRAME — the frame numbers between which the described state actually holds. If the description is \"the car that's turning left\", the interval is the frames in which that car is turning, NOT every frame in which the car is visible. Use the frame numbers printed on the frames. START_FRAME must not be greater than END_FRAME. When MATCHING_IDS is empty, give the full range (the first and last frame numbers); it is ignored.\n\n3. CONFIDENCE — how sure you are of the MATCHING_IDS answer, from 0.0 to 1.0. Be honest: a low number costs the example nothing, and an overconfident wrong answer becomes a wrong label.\n\nRules:\n- Only use numbers that are actually drawn on the frames.\n- Never output coordinates, box values, or pixel positions.\n- Judge motion by comparing the object's position across frames, not by guessing from one frame.\n","user_prompt":"The video has 15 frames, numbered 1 to 15. Each frame is one moment of the clip and has its number burned into the top left corner.\n\nAnswer for each of the following descriptions, in the order given:\n1. the lead cyclist riding in front\n2. the cyclist wearing a bright pink jersey on the right","questions":["the lead cyclist riding in front","the cyclist wearing a bright pink jersey on the right"]},"outputs":{"raw_response":"{\n  \"answers\": [\n    {\n      \"description_index\": 1,\n      \"matching_ids\": [\n        2\n      ],\n      \"start_frame\": 1,\n      \"end_frame\": 15,\n      \"confidence\": 1.0\n    },\n    {\n      \"description_index\": 2,\n      \"matching_ids\": [\n        3\n      ],\n      \"start_frame\": 1,\n      \"end_frame\": 15,\n      \"confidence\": 1.0\n    }\n  ]\n}","raw_answers":[{"description_index":1,"matching_ids":[2],"start_frame":1,"end_frame":15,"confidence":1.0},{"description_index":2,"matching_ids":[3],"start_frame":1,"end_frame":15,"confidence":1.0}],"answers":[{"expression":"the lead cyclist riding in front","matching_presentation_ids":[2],"start_choice_frame":1,"end_choice_frame":15,"confidence":1.0,"self_reported_confidence":1.0},{"expression":"the cyclist wearing a bright pink jersey on the right","matching_presentation_ids":[3],"start_choice_frame":1,"end_choice_frame":15,"confidence":1.0,"self_reported_confidence":1.0}]}},"verify":{"inputs":{"expressions":2,"tracklets":3,"selections":2,"gates":"interval, count, unknown ids, confidence, then geometry (motion)"},"outputs":{"kept":[{"expression":"the lead cyclist riding in front","kind":"motion","is_negative":false,"tracks":[{"timestamps":[0.0,0.25025,0.5005,0.75075,1.001,1.25125,1.5015,1.75175,2.002,2.25225,2.5025,2.75275,3.003,3.25325,3.5035,3.7537499999999997,4.004,4.25425,4.5045,4.75475,5.005,5.25525,5.5055,5.75575,6.006,6.25625,6.5065,6.756749999999999,7.007,7.25725,7.507499999999999,7.75775,8.008,8.25825,8.5085,8.75875,9.009,9.25925,9.5095,9.75975,10.01,10.26025,10.5105,10.76075,11.011,11.26125,11.5115,11.76175,12.012,12.26225,12.5125,12.762749999999999,13.013,13.26325,13.513499999999999,13.76375,14.014],"boxes":[[0.4802083671092987,0.2092592567205429,0.6203125417232513,0.838888868689537],[0.4802083671092987,0.206481471657753,0.6171875447034836,0.8064814358949661],[0.46927085518836975,0.2101851850748062,0.6031250357627869,0.7935184985399246],[0.46145835518836975,0.2203703671693802,0.5921875238418579,0.7879629284143448],[0.45104169845581055,0.23333333432674408,0.5848958790302277,0.8129629343748093],[0.44322919845581055,0.2342592477798462,0.5677083730697632,0.7731481194496155],[0.4437500238418579,0.2370370328426361,0.5744791924953461,0.8055555522441864],[0.4255208671092987,0.23888888955116272,0.555208370089531,0.8166666328907013],[0.4229166805744171,0.24166665971279144,0.5614583492279053,0.7861111015081406],[0.42500001192092896,0.25,0.5635416805744171,0.7749999761581421],[0.43020835518836975,0.2537037134170532,0.5635416954755783,0.8046296238899231],[0.4442708492279053,0.2888888716697693,0.5708333551883698,0.7981480956077576],[0.44843751192092896,0.28703704476356506,0.5723958536982536,0.7583333253860474],[0.4661458432674408,0.25740739703178406,0.5927083492279053,0.7722221910953522],[0.4791666865348816,0.2620370388031006,0.6072916984558105,0.7935184836387634],[0.4739583432674408,0.2638888955116272,0.6015625149011612,0.7833333015441895],[0.48125001788139343,0.2527777850627899,0.6093750298023224,0.7537036836147308],[0.47083336114883423,0.2509259283542633,0.6046875417232513,0.7888889014720917],[0.44166669249534607,0.24444444477558136,0.5812500268220901,0.8138888627290726],[0.44166669249534607,0.23796296119689941,0.5833333730697632,0.7972221970558167],[0.4479166865348816,0.22870369255542755,0.5921875238418579,0.781481459736824],[0.4520833492279053,0.2222222238779068,0.6015625298023224,0.8138888627290726],[0.4687500298023224,0.2231481373310089,0.617708370089531,0.8249999582767487],[0.49427086114883423,0.21666666865348816,0.6421875357627869,0.7861110866069794],[0.5250000357627869,0.206481471657753,0.6781250387430191,0.8074073940515518],[0.5473958849906921,0.20462962985038757,0.7062500566244125,0.8212963044643402],[0.5546875,0.2101851850748062,0.7156250029802322,0.7824074178934097],[0.5463541746139526,0.20277777314186096,0.7156250178813934,0.7824073731899261],[0.46302086114883423,0.2074074000120163,0.7005208730697632,0.8268518298864365],[0.5140625238418579,0.21203702688217163,0.6697916984558105,0.8231481313705444],[0.48645836114883423,0.19166666269302368,0.6453125327825546,0.7842592597007751],[0.4598958492279053,0.32314813137054443,0.6250000298023224,0.824999988079071],[0.4385416805744171,0.32592591643333435,0.6062500178813934,0.8444444239139557],[0.4130208492279053,0.3583333194255829,0.5854166895151138,0.8240740597248077],[0.3968750238418579,0.10185185074806213,0.6364583671092987,0.8240740597248077],[0.3515625298023224,0.18518517911434174,0.5375000387430191,0.8527777642011642],[0.3411458432674408,0.192592591047287,0.525520846247673,0.8398148119449615],[0.3296875059604645,0.1787036955356598,0.5166666805744171,0.8064814507961273],[0.31041666865348816,0.19074073433876038,0.49166667461395264,0.8425925672054291],[0.30520835518836975,0.19722221791744232,0.4854166954755783,0.8462962955236435],[0.3036458492279053,0.20000000298023224,0.4843750298023224,0.8185185343027115],[0.2796875238418579,0.19722221791744232,0.4895833730697632,0.8185185045003891],[0.29635417461395264,0.20277777314186096,0.4609375149011612,0.8444444239139557],[0.31822916865348816,0.2074074000120163,0.48489584028720856,0.8435184806585312],[0.3218750059604645,0.20185184478759766,0.4843750149011612,0.7990740537643433],[0.3213541805744171,0.19814814627170563,0.4843750298023224,0.8305555135011673],[0.30781251192092896,0.20555555820465088,0.4750000238418579,0.8499999642372131],[0.3098958432674408,0.19629628956317902,0.4776041805744171,0.8203703314065933],[0.3072916865348816,0.20185184478759766,0.4765625298023224,0.8064814805984497],[0.32447919249534607,0.18518517911434174,0.5015625357627869,0.8398148268461227],[0.3229166865348816,0.18611110746860504,0.5005208551883698,0.7500000149011612],[0.3239583373069763,0.18888889253139496,0.4932291805744171,0.7907407134771347],[0.31666669249534607,0.1953703612089157,0.47864586114883423,0.8268518298864365],[0.36666667461395264,0.2074074000120163,0.5026041865348816,0.8268518298864365],[0.33020836114883423,0.21759258210659027,0.4869792014360428,0.803703710436821],[0.3307291865348816,0.2203703671693802,0.4812500327825546,0.7888888865709305],[0.3317708373069763,0.22962962090969086,0.47604167461395264,0.7999999970197678]]}]},{"expression":"the cyclist wearing a bright pink jersey on the right","kind":"appearance","is_negative":false,"tracks":[{"timestamps":[0.0,0.25025,0.5005,0.75075,1.001,1.25125,1.5015,1.75175,2.002,2.25225,2.5025,2.75275,3.003,3.25325,3.5035,3.7537499999999997,4.004,4.25425,4.5045,4.75475,5.005,5.25525,5.5055,5.75575,6.25625,7.25725,7.507499999999999,7.75775,8.008,8.25825,8.75875,9.009,9.25925,9.5095,9.75975,10.01,10.26025,10.5105,10.76075,11.011,11.26125,11.5115,11.76175,12.012,12.26225,12.5125,12.762749999999999,13.013,13.26325,13.513499999999999,13.76375,14.014],"boxes":[[0.5869792103767395,0.3722222149372101,0.6781250461935997,0.6888888776302338],[0.5979167222976685,0.36574074625968933,0.6802083924412727,0.6833333373069763],[0.5958333611488342,0.37037035822868347,0.6786458641290665,0.6666666567325592],[0.5963541865348816,0.3759259283542633,0.6750000268220901,0.6583333313465118],[0.5932291746139526,0.3768518567085266,0.6687500104308128,0.6675925850868225],[0.5864583849906921,0.385185182094574,0.6588542237877846,0.5537036955356598],[0.5843750238418579,0.39351850748062134,0.65416669100523,0.6481481194496155],[0.5671875476837158,0.39537036418914795,0.634895883500576,0.6648148000240326],[0.5645833611488342,0.39537036418914795,0.6307291984558105,0.6712962985038757],[0.5630208849906921,0.39907407760620117,0.6260417252779007,0.6518518626689911],[0.5593750476837158,0.4037036895751953,0.6223958879709244,0.6499999761581421],[0.5598958730697632,0.4037036895751953,0.6171875409781933,0.6601851582527161],[0.5588542222976685,0.4055555462837219,0.6098958924412727,0.6499999910593033],[0.5510417222976685,0.412962943315506,0.6072917245328426,0.6398147940635681],[0.565625011920929,0.4444444477558136,0.603645846247673,0.6129629611968994],[0.5640625357627869,0.5435184836387634,0.5750000365078449,0.6527777388691902],[0.5671875476837158,0.47592592239379883,0.5817708820104599,0.5879629552364349],[0.5593750476837158,0.4731481373310089,0.584895882755518,0.6249999850988388],[0.5463541746139526,0.49166664481163025,0.5864583440124989,0.6509259045124054],[0.5385416746139526,0.44999998807907104,0.5864583440124989,0.6518518328666687],[0.5432292222976685,0.4314814805984497,0.5864583924412727,0.635185182094574],[0.5489583611488342,0.4277777671813965,0.5697916951030493,0.5703703612089157],[0.5604166984558105,0.42499998211860657,0.5781250335276127,0.5814814567565918],[0.5770833492279053,0.45092591643333435,0.5968750175088644,0.5305555462837219],[0.5901042222976685,0.5037037134170532,0.6052083894610405,0.5407407507300377],[0.6015625596046448,0.43888887763023376,0.6260417271405458,0.527777761220932],[0.5843750238418579,0.4166666567325592,0.60729169100523,0.5268518403172493],[0.5671875476837158,0.4185185134410858,0.59218754991889,0.6444444358348846],[0.5552083849906921,0.4194444417953491,0.5750000532716513,0.6240740716457367],[0.5416666865348816,0.42129629850387573,0.5921875238418579,0.6546296328306198],[0.5244792103767395,0.4296296238899231,0.5718750469386578,0.635185182094574],[0.5062500238418579,0.4259259104728699,0.55520835891366,0.6518518328666687],[0.31041666865348816,0.4268518388271332,0.5572916865348816,0.6537036895751953],[0.503125011920929,0.4296296238899231,0.5536458492279053,0.635185182094574],[0.46302086114883423,0.4305555522441864,0.5484375357627869,0.6398148089647293],[0.5010417103767395,0.4305555522441864,0.5489583797752857,0.64907406270504],[0.4843750298023224,0.43703702092170715,0.5317708663642406,0.6416666507720947],[0.4843750298023224,0.43981480598449707,0.5307292006909847,0.6342592537403107],[0.5005208849906921,0.43703702092170715,0.5401042215526104,0.6407407224178314],[0.5036458373069763,0.43981480598449707,0.5401041731238365,0.6379629522562027],[0.5041667222976685,0.43981480598449707,0.5395833924412727,0.6212962865829468],[0.5145833492279053,0.4444444477558136,0.5375000163912773,0.635185182094574],[0.5135416984558105,0.4462962746620178,0.5401041992008686,0.5981481224298477],[0.5119792222976685,0.44814813137054443,0.5406250562518835,0.6222222000360489],[0.5026041865348816,0.45092591643333435,0.5364583544433117,0.6222222149372101],[0.499479204416275,0.44814813137054443,0.5333333723247051,0.6277777552604675],[0.4906250238418579,0.4277777671813965,0.5229166932404041,0.6185185015201569],[0.4911458492279053,0.45277777314186096,0.5208333507180214,0.6166666597127914],[0.5,0.45370370149612427,0.5296875014901161,0.6240740716457367],[0.4989583492279053,0.4546296298503876,0.5260416846722364,0.6175925880670547],[0.49635419249534607,0.46296295523643494,0.5244791936129332,0.6064814776182175],[0.49010419845581055,0.460185170173645,0.5151042006909847,0.6166666448116302]]}]}],"rejected":{},"rejection_counts":{},"row":{"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--Dx9HVcQWs_chunk_001_scene_034.mp4","scene_key":"--Dx9HVcQWs_chunk_001_scene_034","category":"test_vehicles","height":1080,"width":1920,"fps":23.976023976023978,"num_frames":338,"duration":14.097416666666666,"rmot_expressions":[{"expression":"the lead cyclist riding in front","kind":"motion","is_negative":false,"tracks":[{"timestamps":[0.0,0.25025,0.5005,0.75075,1.001,1.25125,1.5015,1.75175,2.002,2.25225,2.5025,2.75275,3.003,3.25325,3.5035,3.7537499999999997,4.004,4.25425,4.5045,4.75475,5.005,5.25525,5.5055,5.75575,6.006,6.25625,6.5065,6.756749999999999,7.007,7.25725,7.507499999999999,7.75775,8.008,8.25825,8.5085,8.75875,9.009,9.25925,9.5095,9.75975,10.01,10.26025,10.5105,10.76075,11.011,11.26125,11.5115,11.76175,12.012,12.26225,12.5125,12.762749999999999,13.013,13.26325,13.513499999999999,13.76375,14.014],"boxes":[[0.4802083671092987,0.2092592567205429,0.6203125417232513,0.838888868689537],[0.4802083671092987,0.206481471657753,0.6171875447034836,0.8064814358949661],[0.46927085518836975,0.2101851850748062,0.6031250357627869,0.7935184985399246],[0.46145835518836975,0.2203703671693802,0.5921875238418579,0.7879629284143448],[0.45104169845581055,0.23333333432674408,0.5848958790302277,0.8129629343748093],[0.44322919845581055,0.2342592477798462,0.5677083730697632,0.7731481194496155],[0.4437500238418579,0.2370370328426361,0.5744791924953461,0.8055555522441864],[0.4255208671092987,0.23888888955116272,0.555208370089531,0.8166666328907013],[0.4229166805744171,0.24166665971279144,0.5614583492279053,0.7861111015081406],[0.42500001192092896,0.25,0.5635416805744171,0.7749999761581421],[0.43020835518836975,0.2537037134170532,0.5635416954755783,0.8046296238899231],[0.4442708492279053,0.2888888716697693,0.5708333551883698,0.7981480956077576],[0.44843751192092896,0.28703704476356506,0.5723958536982536,0.7583333253860474],[0.4661458432674408,0.25740739703178406,0.5927083492279053,0.7722221910953522],[0.4791666865348816,0.2620370388031006,0.6072916984558105,0.7935184836387634],[0.4739583432674408,0.2638888955116272,0.6015625149011612,0.7833333015441895],[0.48125001788139343,0.2527777850627899,0.6093750298023224,0.7537036836147308],[0.47083336114883423,0.2509259283542633,0.6046875417232513,0.7888889014720917],[0.44166669249534607,0.24444444477558136,0.5812500268220901,0.8138888627290726],[0.44166669249534607,0.23796296119689941,0.5833333730697632,0.7972221970558167],[0.4479166865348816,0.22870369255542755,0.5921875238418579,0.781481459736824],[0.4520833492279053,0.2222222238779068,0.6015625298023224,0.8138888627290726],[0.4687500298023224,0.2231481373310089,0.617708370089531,0.8249999582767487],[0.49427086114883423,0.21666666865348816,0.6421875357627869,0.7861110866069794],[0.5250000357627869,0.206481471657753,0.6781250387430191,0.8074073940515518],[0.5473958849906921,0.20462962985038757,0.7062500566244125,0.8212963044643402],[0.5546875,0.2101851850748062,0.7156250029802322,0.7824074178934097],[0.5463541746139526,0.20277777314186096,0.7156250178813934,0.7824073731899261],[0.46302086114883423,0.2074074000120163,0.7005208730697632,0.8268518298864365],[0.5140625238418579,0.21203702688217163,0.6697916984558105,0.8231481313705444],[0.48645836114883423,0.19166666269302368,0.6453125327825546,0.7842592597007751],[0.4598958492279053,0.32314813137054443,0.6250000298023224,0.824999988079071],[0.4385416805744171,0.32592591643333435,0.6062500178813934,0.8444444239139557],[0.4130208492279053,0.3583333194255829,0.5854166895151138,0.8240740597248077],[0.3968750238418579,0.10185185074806213,0.6364583671092987,0.8240740597248077],[0.3515625298023224,0.18518517911434174,0.5375000387430191,0.8527777642011642],[0.3411458432674408,0.192592591047287,0.525520846247673,0.8398148119449615],[0.3296875059604645,0.1787036955356598,0.5166666805744171,0.8064814507961273],[0.31041666865348816,0.19074073433876038,0.49166667461395264,0.8425925672054291],[0.30520835518836975,0.19722221791744232,0.4854166954755783,0.8462962955236435],[0.3036458492279053,0.20000000298023224,0.4843750298023224,0.8185185343027115],[0.2796875238418579,0.19722221791744232,0.4895833730697632,0.8185185045003891],[0.29635417461395264,0.20277777314186096,0.4609375149011612,0.8444444239139557],[0.31822916865348816,0.2074074000120163,0.48489584028720856,0.8435184806585312],[0.3218750059604645,0.20185184478759766,0.4843750149011612,0.7990740537643433],[0.3213541805744171,0.19814814627170563,0.4843750298023224,0.8305555135011673],[0.30781251192092896,0.20555555820465088,0.4750000238418579,0.8499999642372131],[0.3098958432674408,0.19629628956317902,0.4776041805744171,0.8203703314065933],[0.3072916865348816,0.20185184478759766,0.4765625298023224,0.8064814805984497],[0.32447919249534607,0.18518517911434174,0.5015625357627869,0.8398148268461227],[0.3229166865348816,0.18611110746860504,0.5005208551883698,0.7500000149011612],[0.3239583373069763,0.18888889253139496,0.4932291805744171,0.7907407134771347],[0.31666669249534607,0.1953703612089157,0.47864586114883423,0.8268518298864365],[0.36666667461395264,0.2074074000120163,0.5026041865348816,0.8268518298864365],[0.33020836114883423,0.21759258210659027,0.4869792014360428,0.803703710436821],[0.3307291865348816,0.2203703671693802,0.4812500327825546,0.7888888865709305],[0.3317708373069763,0.22962962090969086,0.47604167461395264,0.7999999970197678]]}]},{"expression":"the cyclist wearing a bright pink jersey on the right","kind":"appearance","is_negative":false,"tracks":[{"timestamps":[0.0,0.25025,0.5005,0.75075,1.001,1.25125,1.5015,1.75175,2.002,2.25225,2.5025,2.75275,3.003,3.25325,3.5035,3.7537499999999997,4.004,4.25425,4.5045,4.75475,5.005,5.25525,5.5055,5.75575,6.25625,7.25725,7.507499999999999,7.75775,8.008,8.25825,8.75875,9.009,9.25925,9.5095,9.75975,10.01,10.26025,10.5105,10.76075,11.011,11.26125,11.5115,11.76175,12.012,12.26225,12.5125,12.762749999999999,13.013,13.26325,13.513499999999999,13.76375,14.014],"boxes":[[0.5869792103767395,0.3722222149372101,0.6781250461935997,0.6888888776302338],[0.5979167222976685,0.36574074625968933,0.6802083924412727,0.6833333373069763],[0.5958333611488342,0.37037035822868347,0.6786458641290665,0.6666666567325592],[0.5963541865348816,0.3759259283542633,0.6750000268220901,0.6583333313465118],[0.5932291746139526,0.3768518567085266,0.6687500104308128,0.6675925850868225],[0.5864583849906921,0.385185182094574,0.6588542237877846,0.5537036955356598],[0.5843750238418579,0.39351850748062134,0.65416669100523,0.6481481194496155],[0.5671875476837158,0.39537036418914795,0.634895883500576,0.6648148000240326],[0.5645833611488342,0.39537036418914795,0.6307291984558105,0.6712962985038757],[0.5630208849906921,0.39907407760620117,0.6260417252779007,0.6518518626689911],[0.5593750476837158,0.4037036895751953,0.6223958879709244,0.6499999761581421],[0.5598958730697632,0.4037036895751953,0.6171875409781933,0.6601851582527161],[0.5588542222976685,0.4055555462837219,0.6098958924412727,0.6499999910593033],[0.5510417222976685,0.412962943315506,0.6072917245328426,0.6398147940635681],[0.565625011920929,0.4444444477558136,0.603645846247673,0.6129629611968994],[0.5640625357627869,0.5435184836387634,0.5750000365078449,0.6527777388691902],[0.5671875476837158,0.47592592239379883,0.5817708820104599,0.5879629552364349],[0.5593750476837158,0.4731481373310089,0.584895882755518,0.6249999850988388],[0.5463541746139526,0.49166664481163025,0.5864583440124989,0.6509259045124054],[0.5385416746139526,0.44999998807907104,0.5864583440124989,0.6518518328666687],[0.5432292222976685,0.4314814805984497,0.5864583924412727,0.635185182094574],[0.5489583611488342,0.4277777671813965,0.5697916951030493,0.5703703612089157],[0.5604166984558105,0.42499998211860657,0.5781250335276127,0.5814814567565918],[0.5770833492279053,0.45092591643333435,0.5968750175088644,0.5305555462837219],[0.5901042222976685,0.5037037134170532,0.6052083894610405,0.5407407507300377],[0.6015625596046448,0.43888887763023376,0.6260417271405458,0.527777761220932],[0.5843750238418579,0.4166666567325592,0.60729169100523,0.5268518403172493],[0.5671875476837158,0.4185185134410858,0.59218754991889,0.6444444358348846],[0.5552083849906921,0.4194444417953491,0.5750000532716513,0.6240740716457367],[0.5416666865348816,0.42129629850387573,0.5921875238418579,0.6546296328306198],[0.5244792103767395,0.4296296238899231,0.5718750469386578,0.635185182094574],[0.5062500238418579,0.4259259104728699,0.55520835891366,0.6518518328666687],[0.31041666865348816,0.4268518388271332,0.5572916865348816,0.6537036895751953],[0.503125011920929,0.4296296238899231,0.5536458492279053,0.635185182094574],[0.46302086114883423,0.4305555522441864,0.5484375357627869,0.6398148089647293],[0.5010417103767395,0.4305555522441864,0.5489583797752857,0.64907406270504],[0.4843750298023224,0.43703702092170715,0.5317708663642406,0.6416666507720947],[0.4843750298023224,0.43981480598449707,0.5307292006909847,0.6342592537403107],[0.5005208849906921,0.43703702092170715,0.5401042215526104,0.6407407224178314],[0.5036458373069763,0.43981480598449707,0.5401041731238365,0.6379629522562027],[0.5041667222976685,0.43981480598449707,0.5395833924412727,0.6212962865829468],[0.5145833492279053,0.4444444477558136,0.5375000163912773,0.635185182094574],[0.5135416984558105,0.4462962746620178,0.5401041992008686,0.5981481224298477],[0.5119792222976685,0.44814813137054443,0.5406250562518835,0.6222222000360489],[0.5026041865348816,0.45092591643333435,0.5364583544433117,0.6222222149372101],[0.499479204416275,0.44814813137054443,0.5333333723247051,0.6277777552604675],[0.4906250238418579,0.4277777671813965,0.5229166932404041,0.6185185015201569],[0.4911458492279053,0.45277777314186096,0.5208333507180214,0.6166666597127914],[0.5,0.45370370149612427,0.5296875014901161,0.6240740716457367],[0.4989583492279053,0.4546296298503876,0.5260416846722364,0.6175925880670547],[0.49635419249534607,0.46296295523643494,0.5244791936129332,0.6064814776182175],[0.49010419845581055,0.460185170173645,0.5151042006909847,0.6166666448116302]]}]}],"rmot_expression_count":2,"rmot_negative_count":0,"rmot_motion_count":1}}}}},{"index":10,"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_015_scene_004.mp4","status":"complete","steps":{"decode":{"inputs":{"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_015_scene_004.mp4","sample_fps":4.0},"outputs":{"width":1920,"height":1080,"source_fps":50.0,"duration_seconds":20.48,"frames_kept":79,"blank_frames_dropped":0,"timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88,10.14,10.4,10.66,10.92,11.18,11.44,11.700000000000001,11.96,12.22,12.48,12.74,13.0,13.26,13.52,13.780000000000001,14.040000000000001,14.3,14.56,14.82,15.08,15.34,15.6,15.860000000000001,16.12,16.38,16.64,16.9,17.16,17.42,17.68,17.94,18.2,18.46,18.72,18.98,19.240000000000002,19.5,19.76,20.02,20.28],"error":""}},"propose":{"inputs":{"model":"gemini-3.5-flash-lite","temperature":0.4,"video_bytes":32060538,"system_prompt":"You are annotating video clips to build training data for referring multi-object tracking: given a video and a natural-language description, a model must output the boxes of every object matching that description, over time.\n\nYour job is to write the descriptions. You do NOT output coordinates, boxes, pixel positions, or any numbers about where things are. A separate detector finds the objects.\n\nFor every clip you produce a list of items. Each item has two texts that must NOT be the same:\n\nEXPRESSION — what a person would say to single out the object or objects. It may describe motion, temporal order, interaction with other objects, and spatial relations. It should read like something a person would actually say.\n  Good: \"the black car that's turning left\"\n  Good: \"the cyclist who overtakes the bus\"\n  Good: \"the two people who walk out of frame on the right\"\n\nNOUN_PHRASE — the same object as a plain, timeless visual category with its visible attributes, and NOTHING else. The detector is prompted with this text and looks at ONE frame at a time, so anything about time, order, or action is useless to it and makes it fail.\n  Allowed: category nouns, colours, materials, clothing, size, type\n  Forbidden: verbs and participles of action (\"turning\", \"walks\", \"is moving\"), temporal words (\"then\", \"starts\", \"after\", \"first\", \"last\"), relative clauses (\"that ...\", \"which ...\", \"who ...\"), ordinals and rank words (\"first\", \"leftmost\", \"nearest\"), and anything over six words.\n  \"the black car that's turning left\" -> \"black car\"\n  \"the cyclist who overtakes the bus\" -> \"cyclist\"\n  \"the two people who walk out of frame on the right\" -> \"person\"\n\nKIND — what makes the expression hard: \"motion\", \"appearance\", \"spatial\", or \"interaction\". Aim for about 35% \"motion\" items: motion is the part a single-frame model cannot already solve, so it is the most valuable, but a list of only motion items is not usable training data either.\n\nEXPECTED_COUNT — how many distinct objects in the clip match the EXPRESSION. Count carefully; this is checked against the detector's answer, and a wrong count discards the item.\n\nIS_NEGATIVE — at most ONE item may be a negative, and only when you also write at least one non-negative item. A negative is an expression that recombines attributes that ARE present in the clip into a whole that is NOT. If the only black car is moving, \"the parked black car\" is a good negative. Its NOUN_PHRASE must still be the category that IS present (\"black car\"), because that is what makes the absence checkable rather than merely unverified. A negative's EXPECTED_COUNT is 0.\n\nRules that discard an item, so follow them exactly:\n- Never write coordinates, box values, pixel positions, or percentages of the frame.\n- Never let a NOUN_PHRASE contain a verb of action, a temporal word, a relative clause, or an ordinal.\n- Never write an EXPRESSION whose object is not visible in the clip, unless it is marked IS_NEGATIVE.\n- Prefer objects that are clearly visible and identifiable over background clutter.\n- Do not describe the camera or the shot; describe objects in the scene.\n","user_prompt":"Watch this clip and write between 1 and 3 items following the rules. Cover different objects rather than restating one object several ways, and include any negative in the same list.\n"},"outputs":{"raw_response":"{\n  \"items\": [\n    {\n      \"expression\": \"the man with a long beard talking to the camera\",\n      \"noun_phrase\": \"man with a beard\",\n      \"kind\": \"appearance\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the woman standing beside the man\",\n      \"noun_phrase\": \"woman\",\n      \"kind\": \"spatial\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the dog sitting on the ground\",\n      \"noun_phrase\": \"dog\",\n      \"kind\": \"appearance\",\n      \"expected_count\": 0,\n      \"is_negative\": true\n    }\n  ]\n}","raw_items":[{"expression":"the man with a long beard talking to the camera","noun_phrase":"man with a beard","kind":"appearance","expected_count":1,"is_negative":false},{"expression":"the woman standing beside the man","noun_phrase":"woman","kind":"spatial","expected_count":1,"is_negative":false},{"expression":"the dog sitting on the ground","noun_phrase":"dog","kind":"appearance","expected_count":0,"is_negative":true}],"accepted":[{"expression":"the man with a long beard talking to the camera","noun_phrase":"man with a beard","kind":"appearance","expected_count":1,"is_negative":false},{"expression":"the woman standing beside the man","noun_phrase":"woman","kind":"spatial","expected_count":1,"is_negative":false},{"expression":"the dog sitting on the ground","noun_phrase":"dog","kind":"appearance","expected_count":0,"is_negative":true}],"gated":[]}},"track":{"inputs":{"tracker":"Sam31MultiplexTracker","model_id":"/mnt/ceph3/ec/jacky/sam3.1","noun_phrases":["dog","man with a beard","woman"],"expressions_in":3,"frames":79,"sample_fps":4.0,"frame_size":"1920x1080","timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88,10.14,10.4,10.66,10.92,11.18,11.44,11.700000000000001,11.96,12.22,12.48,12.74,13.0,13.26,13.52,13.780000000000001,14.040000000000001,14.3,14.56,14.82,15.08,15.34,15.6,15.860000000000001,16.12,16.38,16.64,16.9,17.16,17.42,17.68,17.94,18.2,18.46,18.72,18.98,19.240000000000002,19.5,19.76,20.02,20.28],"session":"one session over 79 JPEG frames; then per phrase: reset_session, add_prompt(text) on frame 0, propagate_in_video","passes":3},"outputs":{"raw_tracks":[{"noun_phrase":"dog","object_id":0,"observations":[{"timestamp":4.94,"box":[0.35625001788139343,0.7574073672294617,0.38750001788139343,0.7935184761881828],"occluded":false},{"timestamp":5.2,"box":[0.36250001192092896,0.7527777552604675,0.39791668206453323,0.7861110866069794],"occluded":false},{"timestamp":5.46,"box":[0.35364586114883423,0.7407407164573669,0.3942708633840084,0.7796296030282974],"occluded":false},{"timestamp":5.72,"box":[0.35885417461395264,0.7342592477798462,0.39010417461395264,0.7768518403172493],"occluded":false},{"timestamp":5.98,"box":[0.36770835518836975,0.7518518567085266,0.38802085630595684,0.7833333387970924],"occluded":false},{"timestamp":6.5,"box":[0.3541666865348816,0.7277777791023254,0.3817708548158407,0.7750000022351742],"occluded":false},{"timestamp":6.76,"box":[0.3463541865348816,0.7175925970077515,0.37500002048909664,0.7675925977528095],"occluded":false},{"timestamp":7.28,"box":[0.30885419249534607,0.7240740656852722,0.3385416939854622,0.7611111029982567],"occluded":false},{"timestamp":11.700000000000001,"box":[0.3109375238418579,0.7268518209457397,0.35572919249534607,0.7759258933365345],"occluded":false},{"timestamp":11.96,"box":[0.30416667461395264,0.7268518209457397,0.3182291751727462,0.7694444134831429],"occluded":false},{"timestamp":12.22,"box":[0.3062500059604645,0.7157407402992249,0.31822917331010103,0.7546296268701553],"occluded":false},{"timestamp":12.48,"box":[0.3192708492279053,0.7175925970077515,0.33385418355464935,0.7555555589497089],"occluded":false},{"timestamp":12.74,"box":[0.3333333432674408,0.7129629254341125,0.3458333443850279,0.7453703321516514],"occluded":false},{"timestamp":13.0,"box":[0.3447916805744171,0.6990740895271301,0.35625001415610313,0.7388889044523239],"occluded":false},{"timestamp":13.52,"box":[0.3270833492279053,0.7064814567565918,0.3354166829958558,0.7462962716817856],"occluded":false}],"scores":[0.7928802371025085,0.7928802371025085,0.7928802371025085,0.7928802371025085,0.7928802371025085,0.7928802371025085,0.7928802371025085,0.7928802371025085,0.7928802371025085,0.7928802371025085,0.7928802371025085,0.7928802371025085,0.7928802371025085,0.7928802371025085,0.7928802371025085]},{"noun_phrase":"man with a beard","object_id":0,"observations":[{"timestamp":0.0,"box":[0.5562500357627869,0.08148147910833359,0.9994792342185974,0.9703703746199608],"occluded":false},{"timestamp":0.26,"box":[0.3437500298023224,0.20462962985038757,0.9708333909511566,0.9712962806224823],"occluded":false},{"timestamp":0.52,"box":[0.3687500059604645,0.1787036955356598,0.999479204416275,0.9722221791744232],"occluded":false},{"timestamp":0.78,"box":[0.4453125298023224,0.06759259104728699,0.999479204416275,0.9712962806224823],"occluded":false},{"timestamp":1.04,"box":[0.4515625238418579,0.10740740597248077,0.9885417222976685,0.9722222238779068],"occluded":false},{"timestamp":1.3,"box":[0.4401041865348816,0.1101851835846901,0.9708333611488342,0.9722221866250038],"occluded":false},{"timestamp":1.56,"box":[0.4609375298023224,0.04537037014961243,0.999479204416275,0.9722221791744232],"occluded":false},{"timestamp":1.82,"box":[0.45729169249534607,0.13518518209457397,0.9895834028720856,0.9722222089767456],"occluded":false},{"timestamp":2.08,"box":[0.45625001192092896,0.1944444477558136,0.9781250357627869,0.972222238779068],"occluded":false},{"timestamp":2.34,"box":[0.45468753576278687,0.10833333432674408,0.97552090883255,0.9722221940755844],"occluded":false},{"timestamp":2.6,"box":[0.4505208432674408,0.09259258955717087,0.9708333909511566,0.972222201526165],"occluded":false},{"timestamp":2.86,"box":[0.4437500238418579,0.13611111044883728,0.9729167222976685,0.9722221791744232],"occluded":false},{"timestamp":3.12,"box":[0.44166669249534607,0.1537037044763565,0.9828125536441803,0.9722222238779068],"occluded":false},{"timestamp":3.38,"box":[0.45625001192092896,0.12314814329147339,0.9994792342185974,0.9722222089767456],"occluded":false},{"timestamp":3.64,"box":[0.45781251788139343,0.12870369851589203,0.999479204416275,0.9722221940755844],"occluded":false},{"timestamp":3.9,"box":[0.4593750238418579,0.12870369851589203,0.9994792342185974,0.9722221940755844],"occluded":false},{"timestamp":4.16,"box":[0.4598958492279053,0.15833333134651184,0.9994792342185974,0.9722221791744232],"occluded":false},{"timestamp":4.42,"box":[0.4583333432674408,0.11574073880910873,0.999479204416275,0.9722222313284874],"occluded":false},{"timestamp":4.68,"box":[0.4453125298023224,0.09166666865348816,0.999479204416275,0.9722221791744232],"occluded":false},{"timestamp":4.94,"box":[0.4375000298023224,0.13148148357868195,0.999479204416275,0.9722222238779068],"occluded":false},{"timestamp":5.2,"box":[0.4479166865348816,0.09444444626569748,0.9994792342185974,0.972222201526165],"occluded":false},{"timestamp":5.46,"box":[0.44583335518836975,0.07129629701375961,0.999479204416275,0.9722222313284874],"occluded":false},{"timestamp":5.72,"box":[0.45104169845581055,0.10833333432674408,0.9994792342185974,0.9722221940755844],"occluded":false},{"timestamp":5.98,"box":[0.43281251192092896,0.13240741193294525,0.9937500357627869,0.9722221940755844],"occluded":false},{"timestamp":6.24,"box":[0.4322916865348816,0.13333332538604736,0.9994792342185974,0.9722222089767456],"occluded":false},{"timestamp":6.5,"box":[0.43437501788139343,0.1537037044763565,0.999479204416275,0.9722222238779068],"occluded":false},{"timestamp":6.76,"box":[0.42760419845581055,0.12870369851589203,0.9994792342185974,0.9722221940755844],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.44583335518836975,0.14722222089767456,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":7.28,"box":[0.4541666805744171,0.13611111044883728,0.999479204416275,0.9722221791744232],"occluded":false},{"timestamp":7.54,"box":[0.45625001192092896,0.09907407313585281,0.9994792342185974,0.9712962582707405],"occluded":false},{"timestamp":7.8,"box":[0.4593750238418579,0.11296296119689941,0.9994792342185974,0.9712962508201599],"occluded":false},{"timestamp":8.06,"box":[0.4598958492279053,0.08981481194496155,0.9994792342185974,0.9712962806224823],"occluded":false},{"timestamp":8.32,"box":[0.4661458432674408,0.13703703880310059,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":8.58,"box":[0.4619791805744171,0.09351851791143417,0.999479204416275,0.9712962731719017],"occluded":false},{"timestamp":8.84,"box":[0.4598958492279053,0.09351851791143417,0.9994792342185974,0.9722222313284874],"occluded":false},{"timestamp":9.1,"box":[0.4557291865348816,0.08703703433275223,0.9994792342185974,0.9722222164273262],"occluded":false},{"timestamp":9.36,"box":[0.45104169845581055,0.07870370149612427,0.9994792342185974,0.9722222089767456],"occluded":false},{"timestamp":9.620000000000001,"box":[0.38333335518836975,0.14999999105930328,0.999479204416275,0.9722222238779068],"occluded":false},{"timestamp":9.88,"box":[0.37552085518836975,0.15648147463798523,0.999479204416275,0.9722221791744232],"occluded":false},{"timestamp":10.14,"box":[0.4114583432674408,0.138888880610466,0.9968750178813934,0.9722221940755844],"occluded":false},{"timestamp":10.4,"box":[0.4375000298023224,0.16018518805503845,0.999479204416275,0.972222238779068],"occluded":false},{"timestamp":10.66,"box":[0.4489583671092987,0.20000000298023224,0.999479204416275,0.9722222238779068],"occluded":false},{"timestamp":10.92,"box":[0.45468753576278687,0.14814814925193787,0.9994792342185974,0.9722221791744232],"occluded":false},{"timestamp":11.18,"box":[0.43697917461395264,0.1379629671573639,0.9994791746139526,0.972222238779068],"occluded":false},{"timestamp":11.44,"box":[0.4520833492279053,0.14166666567325592,0.9994792342185974,0.9722222238779068],"occluded":false},{"timestamp":11.700000000000001,"box":[0.4479166865348816,0.14722222089767456,0.9994792342185974,0.9722222089767456],"occluded":false},{"timestamp":11.96,"box":[0.44583335518836975,0.16944444179534912,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":12.22,"box":[0.46406251192092896,0.15648147463798523,0.9994792342185974,0.9722221791744232],"occluded":false},{"timestamp":12.48,"box":[0.4598958492279053,0.13703703880310059,0.9994792342185974,0.9722222089767456],"occluded":false},{"timestamp":12.74,"box":[0.4505208432674408,0.14074073731899261,0.999479204416275,0.9722221940755844],"occluded":false},{"timestamp":13.0,"box":[0.45468753576278687,0.12777777016162872,0.9994792342185974,0.9722222238779068],"occluded":false},{"timestamp":13.26,"box":[0.46562501788139343,0.13333332538604736,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":13.52,"box":[0.4515625238418579,0.11481481045484543,0.9994792342185974,0.972222201526165],"occluded":false},{"timestamp":13.780000000000001,"box":[0.3760416805744171,0.07129629701375961,0.999479204416275,0.9722222313284874],"occluded":false},{"timestamp":14.040000000000001,"box":[0.35104167461395264,0.06481481343507767,0.9994792342185974,0.9722222164273262],"occluded":false},{"timestamp":14.3,"box":[0.47812503576278687,0.05925925821065903,0.9994792342185974,0.9722222313284874],"occluded":false},{"timestamp":14.56,"box":[0.48229169845581055,0.0824074074625969,0.9994792342185974,0.972222201526165],"occluded":false},{"timestamp":14.82,"box":[0.483854204416275,0.10277777910232544,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":15.08,"box":[0.4755208492279053,0.13055555522441864,0.9994792342185974,0.9722221940755844],"occluded":false},{"timestamp":15.34,"box":[0.452604204416275,0.14444443583488464,0.999479204416275,0.9722221791744232],"occluded":false},{"timestamp":15.6,"box":[0.4583333432674408,0.15833333134651184,0.999479204416275,0.9722221791744232],"occluded":false},{"timestamp":15.860000000000001,"box":[0.45468753576278687,0.1527777761220932,0.9994792342185974,0.9712962955236435],"occluded":false},{"timestamp":16.12,"box":[0.46562501788139343,0.16296295821666718,0.999479204416275,0.9712962955236435],"occluded":false},{"timestamp":16.38,"box":[0.46562501788139343,0.15833333134651184,0.999479204416275,0.9712962806224823],"occluded":false},{"timestamp":16.64,"box":[0.45885419845581055,0.15833333134651184,0.9994792342185974,0.9722221791744232],"occluded":false},{"timestamp":16.9,"box":[0.4645833671092987,0.16018518805503845,0.999479204416275,0.9712962806224823],"occluded":false},{"timestamp":17.16,"box":[0.46927085518836975,0.16388888657093048,0.999479204416275,0.9722222238779068],"occluded":false},{"timestamp":17.42,"box":[0.4755208492279053,0.15648147463798523,0.9994792342185974,0.9722221791744232],"occluded":false},{"timestamp":17.68,"box":[0.47864586114883423,0.14999999105930328,0.9994792342185974,0.9722222238779068],"occluded":false},{"timestamp":17.94,"box":[0.47968751192092896,0.15555554628372192,0.9994792342185974,0.9712962508201599],"occluded":false},{"timestamp":18.2,"box":[0.46250003576278687,0.12314814329147339,0.9994792342185974,0.9712962508201599],"occluded":false},{"timestamp":18.46,"box":[0.3348958492279053,0.03703703731298447,0.9994792342185974,0.9712962731719017],"occluded":false},{"timestamp":18.72,"box":[0.4114583432674408,0.031481482088565826,0.999479204416275,0.9722221866250038],"occluded":false},{"timestamp":18.98,"box":[0.43958336114883423,0.03333333134651184,0.9994792342185974,0.9712962806224823],"occluded":false},{"timestamp":19.240000000000002,"box":[0.43177086114883423,0.09259258955717087,0.9994792342185974,0.9712963029742241],"occluded":false},{"timestamp":19.5,"box":[0.4359375238418579,0.14074073731899261,0.9994792342185974,0.9722221940755844],"occluded":false},{"timestamp":19.76,"box":[0.4333333671092987,0.18425925076007843,0.999479204416275,0.9722222238779068],"occluded":false},{"timestamp":20.02,"box":[0.4442708492279053,0.18796296417713165,0.9994792342185974,0.9722222238779068],"occluded":false},{"timestamp":20.28,"box":[0.4567708671092987,0.17314814031124115,0.999479204416275,0.9722221940755844],"occluded":false}],"scores":[0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917,0.9726027250289917]},{"noun_phrase":"woman","object_id":0,"observations":[{"timestamp":0.0,"box":[0.0,0.23240740597248077,0.25,0.9712962657213211],"occluded":false},{"timestamp":0.26,"box":[0.0,0.1953703612089157,0.27812501788139343,0.9722221940755844],"occluded":false},{"timestamp":0.52,"box":[0.0,0.2203703671693802,0.3786458671092987,0.9749999791383743],"occluded":false},{"timestamp":0.78,"box":[0.0,0.25833332538604736,0.42656251788139343,0.9712962508201599],"occluded":false},{"timestamp":1.04,"box":[0.04270833730697632,0.33981481194496155,0.41822919249534607,0.9712962806224823],"occluded":false},{"timestamp":1.3,"box":[0.0,0.3712962865829468,0.3942708671092987,0.9731481075286865],"occluded":false},{"timestamp":1.56,"box":[0.0,0.364814817905426,0.37708336114883423,0.9722222089767456],"occluded":false},{"timestamp":1.82,"box":[0.0,0.3083333373069763,0.3656250238418579,0.9731481671333313],"occluded":false},{"timestamp":2.08,"box":[0.0,0.2657407224178314,0.36666667461395264,0.9731481373310089],"occluded":false},{"timestamp":2.34,"box":[0.0,0.22407406568527222,0.9395833611488342,0.9722222089767456],"occluded":false},{"timestamp":2.6,"box":[0.0,0.14814814925193787,0.4401041865348816,0.9722221791744232],"occluded":false},{"timestamp":2.86,"box":[0.0,0.18425925076007843,0.4479166865348816,0.9740740805864334],"occluded":false},{"timestamp":3.12,"box":[0.00729166716337204,0.2527777850627899,0.45364584773778915,0.972222238779068],"occluded":false},{"timestamp":3.38,"box":[0.0,0.2925925850868225,0.4322916865348816,0.9731481075286865],"occluded":false},{"timestamp":3.64,"box":[0.0,0.3157407343387604,0.40781253576278687,0.9731481373310089],"occluded":false},{"timestamp":3.9,"box":[0.0,0.3166666626930237,0.3661458492279053,0.9740740656852722],"occluded":false},{"timestamp":4.16,"box":[0.0,0.31296294927597046,0.35208335518836975,0.9722222089767456],"occluded":false},{"timestamp":4.42,"box":[0.0,0.3166666626930237,0.3604166805744171,0.9731481075286865],"occluded":false},{"timestamp":4.68,"box":[0.0,0.2898148000240326,0.3411458432674408,0.9722221791744232],"occluded":false},{"timestamp":4.94,"box":[0.0,0.2787036895751953,0.35364586114883423,0.9722222089767456],"occluded":false},{"timestamp":5.2,"box":[0.0,0.26851850748062134,0.3682291805744171,0.9722222089767456],"occluded":false},{"timestamp":5.46,"box":[0.0,0.2666666507720947,0.39635419845581055,0.9722222089767456],"occluded":false},{"timestamp":5.72,"box":[0.0,0.29629629850387573,0.3916666805744171,0.9740740656852722],"occluded":false},{"timestamp":5.98,"box":[0.0,0.3018518388271332,0.41822919249534607,0.9712962806224823],"occluded":false},{"timestamp":6.24,"box":[0.0,0.34166666865348816,0.43020835518836975,0.9712962806224823],"occluded":false},{"timestamp":6.5,"box":[0.0,0.32407405972480774,0.37968751788139343,0.9722221791744232],"occluded":false},{"timestamp":6.76,"box":[0.0,0.031481482088565826,0.3526041805744171,0.9750000014901161],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.0,0.031481482088565826,0.48072919249534607,0.9722221866250038],"occluded":false},{"timestamp":7.28,"box":[0.0,0.17129629850387573,0.3802083432674408,0.9712963104248047],"occluded":false},{"timestamp":7.54,"box":[0.0,0.1675925850868225,0.41354167461395264,0.9731481075286865],"occluded":false},{"timestamp":7.8,"box":[0.0,0.1675925850868225,0.40572917461395264,0.9722222089767456],"occluded":false},{"timestamp":8.06,"box":[0.0,0.17129629850387573,0.37812501192092896,0.9722222089767456],"occluded":false},{"timestamp":8.32,"box":[0.0,0.25462961196899414,0.3802083432674408,0.9722222089767456],"occluded":false},{"timestamp":8.58,"box":[0.0,0.2648148238658905,0.39270836114883423,0.9990740716457367],"occluded":false},{"timestamp":8.84,"box":[0.0,0.24629628658294678,0.3838541805744171,0.9990740418434143],"occluded":false},{"timestamp":9.1,"box":[0.006770833861082792,0.21574074029922485,0.3802083577029407,0.9722222089767456],"occluded":false},{"timestamp":9.36,"box":[0.0,0.21481481194496155,0.3802083432674408,0.9722221791744232],"occluded":false},{"timestamp":9.620000000000001,"box":[0.0,0.21666666865348816,0.39114585518836975,0.9712962806224823],"occluded":false},{"timestamp":9.88,"box":[0.0,0.18796296417713165,0.39375001192092896,0.9712962657213211],"occluded":false},{"timestamp":10.14,"box":[0.0,0.15555554628372192,0.3776041865348816,0.9990740418434143],"occluded":false},{"timestamp":10.4,"box":[0.0,0.1675925850868225,0.3895833492279053,0.9722222089767456],"occluded":false},{"timestamp":10.66,"box":[0.0,0.17037037014961243,0.3958333432674408,0.9722221791744232],"occluded":false},{"timestamp":10.92,"box":[0.0,0.1675925850868225,0.3734375238418579,0.9712962508201599],"occluded":false},{"timestamp":11.18,"box":[0.0,0.19629628956317902,0.34427085518836975,0.9722222238779068],"occluded":false},{"timestamp":11.44,"box":[0.0,0.2083333283662796,0.3265625238418579,0.9722222238779068],"occluded":false},{"timestamp":11.700000000000001,"box":[0.0,0.18425925076007843,0.32604169845581055,0.9712962657213211],"occluded":false},{"timestamp":11.96,"box":[0.0,0.20185184478759766,0.34947919845581055,0.9722222089767456],"occluded":false},{"timestamp":12.22,"box":[0.0,0.17037037014961243,0.36770835518836975,0.9712962806224823],"occluded":false},{"timestamp":12.48,"box":[0.0,0.05648148059844971,0.4166666865348816,0.9722222089767456],"occluded":false},{"timestamp":12.74,"box":[0.0,0.16296295821666718,0.4218750298023224,0.9722221940755844],"occluded":false},{"timestamp":13.0,"box":[0.0,0.1537037044763565,0.4208333492279053,0.9722222238779068],"occluded":false},{"timestamp":13.26,"box":[0.0,0.18611110746860504,0.43177086114883423,0.9722222238779068],"occluded":false},{"timestamp":13.52,"box":[0.0,0.2074074000120163,0.41875001788139343,0.9712962955236435],"occluded":false},{"timestamp":13.780000000000001,"box":[0.0,0.20462962985038757,0.43958336114883423,0.9712962806224823],"occluded":false},{"timestamp":14.040000000000001,"box":[0.0,0.20277777314186096,0.4208333492279053,0.9712962806224823],"occluded":false},{"timestamp":14.3,"box":[0.0,0.21388888359069824,0.3864583671092987,0.9722222089767456],"occluded":false},{"timestamp":14.56,"box":[0.0,0.21851851046085358,0.3854166865348816,0.9712962657213211],"occluded":false},{"timestamp":14.82,"box":[0.0,0.182407408952713,0.39791667461395264,0.972222238779068],"occluded":false},{"timestamp":15.08,"box":[0.0,0.23148147761821747,0.3958333432674408,0.9722221940755844],"occluded":false},{"timestamp":15.34,"box":[0.0,0.2370370328426361,0.41822919249534607,0.9722221791744232],"occluded":false},{"timestamp":15.6,"box":[0.0,0.24444444477558136,0.43281251192092896,0.9722222238779068],"occluded":false},{"timestamp":15.860000000000001,"box":[0.0,0.2083333283662796,0.42500001192092896,0.9722222238779068],"occluded":false},{"timestamp":16.12,"box":[0.0,0.17499999701976776,0.4463541805744171,0.9722221940755844],"occluded":false},{"timestamp":16.38,"box":[0.0,0.17685185372829437,0.44583335518836975,0.9722221940755844],"occluded":false},{"timestamp":16.64,"box":[0.0,0.12314814329147339,0.44218751788139343,0.9722222089767456],"occluded":false},{"timestamp":16.9,"box":[0.0,0.17407406866550446,0.643750011920929,0.9712962657213211],"occluded":false},{"timestamp":17.16,"box":[0.0031250002793967724,0.2527777850627899,0.42916669277474284,0.972222238779068],"occluded":false},{"timestamp":17.42,"box":[0.0,0.25,0.4192708432674408,0.9712963104248047],"occluded":false},{"timestamp":17.68,"box":[0.0,0.2370370328426361,0.3994791805744171,0.9722221791744232],"occluded":false},{"timestamp":17.94,"box":[0.0,0.1935185194015503,0.36406251788139343,0.9712963104248047],"occluded":false},{"timestamp":18.2,"box":[0.0,0.05092592537403107,0.3906250298023224,0.9722222238779068],"occluded":false},{"timestamp":18.46,"box":[0.0,0.1259259283542633,0.40833336114883423,0.972222238779068],"occluded":false},{"timestamp":18.72,"box":[0.0,0.13333332538604736,0.43802085518836975,0.9712962508201599],"occluded":false},{"timestamp":18.98,"box":[0.0,0.12314814329147339,0.44687503576278687,0.9722222089767456],"occluded":false},{"timestamp":19.240000000000002,"box":[0.0,0.13425925374031067,0.46302086114883423,0.9722221791744232],"occluded":false},{"timestamp":19.5,"box":[0.0,0.1379629671573639,0.4515625238418579,0.972222238779068],"occluded":false},{"timestamp":19.76,"box":[0.0,0.138888880610466,0.45364585518836975,0.9722221940755844],"occluded":false},{"timestamp":20.02,"box":[0.0,0.1787036955356598,0.4609375298023224,0.9722221791744232],"occluded":false},{"timestamp":20.28,"box":[0.0,0.2231481373310089,0.47083336114883423,0.9722221791744232],"occluded":false}],"scores":[0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448,0.9375000596046448]},{"noun_phrase":"woman","object_id":1,"observations":[{"timestamp":3.38,"box":[0.45364585518836975,0.6796296238899231,0.4734375234693289,0.7787036970257759],"occluded":false},{"timestamp":3.64,"box":[0.3895833492279053,0.6916666626930237,0.4114583507180214,0.7944444417953491],"occluded":false},{"timestamp":3.9,"box":[0.34843751788139343,0.7037037014961243,0.3734375201165676,0.8101851791143417],"occluded":false},{"timestamp":4.16,"box":[0.3270833492279053,0.7018518447875977,0.34947918355464935,0.8092592507600784],"occluded":false},{"timestamp":4.42,"box":[0.31041666865348816,0.7092592716217041,0.33489583618938923,0.8185185268521309],"occluded":false},{"timestamp":4.68,"box":[0.30520835518836975,0.6953703761100769,0.3286458570510149,0.8046296313405037],"occluded":false},{"timestamp":4.94,"box":[0.32864585518836975,0.6796296238899231,0.3520833570510149,0.7851851806044579],"occluded":false},{"timestamp":5.2,"box":[0.33906251192092896,0.6722221970558167,0.3625000137835741,0.7787036746740341],"occluded":false},{"timestamp":5.46,"box":[0.35104167461395264,0.6722221970558167,0.3697916753590107,0.7657407149672508],"occluded":false},{"timestamp":5.72,"box":[0.3609375059604645,0.6638888716697693,0.3802083395421505,0.7388888671994209],"occluded":false},{"timestamp":6.5,"box":[0.34166669845581055,0.654629647731781,0.37656253203749657,0.7481481656432152],"occluded":false},{"timestamp":6.76,"box":[0.34791669249534607,0.6462962627410889,0.3651041928678751,0.7249999642372131],"occluded":false},{"timestamp":7.28,"box":[0.29739585518836975,0.645370364189148,0.33020835742354393,0.7194444388151169],"occluded":false}],"scores":[0.6683937907218933,0.6683937907218933,0.6683937907218933,0.6683937907218933,0.6683937907218933,0.6683937907218933,0.6683937907218933,0.6683937907218933,0.6683937907218933,0.6683937907218933,0.6683937907218933,0.6683937907218933,0.6683937907218933]}],"tracklets":[{"noun_phrase":"woman","object_id":1,"presentation_id":1,"observations":[{"timestamp":3.38,"box":[0.45364585518836975,0.6796296238899231,0.4734375234693289,0.7787036970257759],"occluded":false},{"timestamp":3.64,"box":[0.3895833492279053,0.6916666626930237,0.4114583507180214,0.7944444417953491],"occluded":false},{"timestamp":3.9,"box":[0.34843751788139343,0.7037037014961243,0.3734375201165676,0.8101851791143417],"occluded":false},{"timestamp":4.16,"box":[0.3270833492279053,0.7018518447875977,0.34947918355464935,0.8092592507600784],"occluded":false},{"timestamp":4.42,"box":[0.31041666865348816,0.7092592716217041,0.33489583618938923,0.8185185268521309],"occluded":false},{"timestamp":4.68,"box":[0.30520835518836975,0.6953703761100769,0.3286458570510149,0.8046296313405037],"occluded":false},{"timestamp":4.94,"box":[0.32864585518836975,0.6796296238899231,0.3520833570510149,0.7851851806044579],"occluded":false},{"timestamp":5.2,"box":[0.33906251192092896,0.6722221970558167,0.3625000137835741,0.7787036746740341],"occluded":false},{"timestamp":5.46,"box":[0.35104167461395264,0.6722221970558167,0.3697916753590107,0.7657407149672508],"occluded":false},{"timestamp":5.72,"box":[0.3609375059604645,0.6638888716697693,0.3802083395421505,0.7388888671994209],"occluded":false},{"timestamp":6.5,"box":[0.34166669845581055,0.654629647731781,0.37656253203749657,0.7481481656432152],"occluded":false},{"timestamp":6.76,"box":[0.34791669249534607,0.6462962627410889,0.3651041928678751,0.7249999642372131],"occluded":false},{"timestamp":7.28,"box":[0.29739585518836975,0.645370364189148,0.33020835742354393,0.7194444388151169],"occluded":false}],"mean_score":0.6683937907218933},{"noun_phrase":"man with a beard","object_id":0,"presentation_id":2,"observations":[{"timestamp":0.0,"box":[0.5562500357627869,0.08148147910833359,0.9994792342185974,0.9703703746199608],"occluded":false},{"timestamp":0.26,"box":[0.3437500298023224,0.20462962985038757,0.9708333909511566,0.9712962806224823],"occluded":false},{"timestamp":0.52,"box":[0.3687500059604645,0.1787036955356598,0.999479204416275,0.9722221791744232],"occluded":false},{"timestamp":0.78,"box":[0.4453125298023224,0.06759259104728699,0.999479204416275,0.9712962806224823],"occluded":false},{"timestamp":1.04,"box":[0.4515625238418579,0.10740740597248077,0.9885417222976685,0.9722222238779068],"occluded":false},{"timestamp":1.3,"box":[0.4401041865348816,0.1101851835846901,0.9708333611488342,0.9722221866250038],"occluded":false},{"timestamp":1.56,"box":[0.4609375298023224,0.04537037014961243,0.999479204416275,0.9722221791744232],"occluded":false},{"timestamp":1.82,"box":[0.45729169249534607,0.13518518209457397,0.9895834028720856,0.9722222089767456],"occluded":false},{"timestamp":2.08,"box":[0.45625001192092896,0.1944444477558136,0.9781250357627869,0.972222238779068],"occluded":false},{"timestamp":2.34,"box":[0.45468753576278687,0.10833333432674408,0.97552090883255,0.9722221940755844],"occluded":false},{"timestamp":2.6,"box":[0.4505208432674408,0.09259258955717087,0.9708333909511566,0.972222201526165],"occluded":false},{"timestamp":2.86,"box":[0.4437500238418579,0.13611111044883728,0.9729167222976685,0.9722221791744232],"occluded":false},{"timestamp":3.12,"box":[0.44166669249534607,0.1537037044763565,0.9828125536441803,0.9722222238779068],"occluded":false},{"timestamp":3.38,"box":[0.45625001192092896,0.12314814329147339,0.9994792342185974,0.9722222089767456],"occluded":false},{"timestamp":3.64,"box":[0.45781251788139343,0.12870369851589203,0.999479204416275,0.9722221940755844],"occluded":false},{"timestamp":3.9,"box":[0.4593750238418579,0.12870369851589203,0.9994792342185974,0.9722221940755844],"occluded":false},{"timestamp":4.16,"box":[0.4598958492279053,0.15833333134651184,0.9994792342185974,0.9722221791744232],"occluded":false},{"timestamp":4.42,"box":[0.4583333432674408,0.11574073880910873,0.999479204416275,0.9722222313284874],"occluded":false},{"timestamp":4.68,"box":[0.4453125298023224,0.09166666865348816,0.999479204416275,0.9722221791744232],"occluded":false},{"timestamp":4.94,"box":[0.4375000298023224,0.13148148357868195,0.999479204416275,0.9722222238779068],"occluded":false},{"timestamp":5.2,"box":[0.4479166865348816,0.09444444626569748,0.9994792342185974,0.972222201526165],"occluded":false},{"timestamp":5.46,"box":[0.44583335518836975,0.07129629701375961,0.999479204416275,0.9722222313284874],"occluded":false},{"timestamp":5.72,"box":[0.45104169845581055,0.10833333432674408,0.9994792342185974,0.9722221940755844],"occluded":false},{"timestamp":5.98,"box":[0.43281251192092896,0.13240741193294525,0.9937500357627869,0.9722221940755844],"occluded":false},{"timestamp":6.24,"box":[0.4322916865348816,0.13333332538604736,0.9994792342185974,0.9722222089767456],"occluded":false},{"timestamp":6.5,"box":[0.43437501788139343,0.1537037044763565,0.999479204416275,0.9722222238779068],"occluded":false},{"timestamp":6.76,"box":[0.42760419845581055,0.12870369851589203,0.9994792342185974,0.9722221940755844],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.44583335518836975,0.14722222089767456,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":7.28,"box":[0.4541666805744171,0.13611111044883728,0.999479204416275,0.9722221791744232],"occluded":false},{"timestamp":7.54,"box":[0.45625001192092896,0.09907407313585281,0.9994792342185974,0.9712962582707405],"occluded":false},{"timestamp":7.8,"box":[0.4593750238418579,0.11296296119689941,0.9994792342185974,0.9712962508201599],"occluded":false},{"timestamp":8.06,"box":[0.4598958492279053,0.08981481194496155,0.9994792342185974,0.9712962806224823],"occluded":false},{"timestamp":8.32,"box":[0.4661458432674408,0.13703703880310059,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":8.58,"box":[0.4619791805744171,0.09351851791143417,0.999479204416275,0.9712962731719017],"occluded":false},{"timestamp":8.84,"box":[0.4598958492279053,0.09351851791143417,0.9994792342185974,0.9722222313284874],"occluded":false},{"timestamp":9.1,"box":[0.4557291865348816,0.08703703433275223,0.9994792342185974,0.9722222164273262],"occluded":false},{"timestamp":9.36,"box":[0.45104169845581055,0.07870370149612427,0.9994792342185974,0.9722222089767456],"occluded":false},{"timestamp":9.620000000000001,"box":[0.38333335518836975,0.14999999105930328,0.999479204416275,0.9722222238779068],"occluded":false},{"timestamp":9.88,"box":[0.37552085518836975,0.15648147463798523,0.999479204416275,0.9722221791744232],"occluded":false},{"timestamp":10.14,"box":[0.4114583432674408,0.138888880610466,0.9968750178813934,0.9722221940755844],"occluded":false},{"timestamp":10.4,"box":[0.4375000298023224,0.16018518805503845,0.999479204416275,0.972222238779068],"occluded":false},{"timestamp":10.66,"box":[0.4489583671092987,0.20000000298023224,0.999479204416275,0.9722222238779068],"occluded":false},{"timestamp":10.92,"box":[0.45468753576278687,0.14814814925193787,0.9994792342185974,0.9722221791744232],"occluded":false},{"timestamp":11.18,"box":[0.43697917461395264,0.1379629671573639,0.9994791746139526,0.972222238779068],"occluded":false},{"timestamp":11.44,"box":[0.4520833492279053,0.14166666567325592,0.9994792342185974,0.9722222238779068],"occluded":false},{"timestamp":11.700000000000001,"box":[0.4479166865348816,0.14722222089767456,0.9994792342185974,0.9722222089767456],"occluded":false},{"timestamp":11.96,"box":[0.44583335518836975,0.16944444179534912,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":12.22,"box":[0.46406251192092896,0.15648147463798523,0.9994792342185974,0.9722221791744232],"occluded":false},{"timestamp":12.48,"box":[0.4598958492279053,0.13703703880310059,0.9994792342185974,0.9722222089767456],"occluded":false},{"timestamp":12.74,"box":[0.4505208432674408,0.14074073731899261,0.999479204416275,0.9722221940755844],"occluded":false},{"timestamp":13.0,"box":[0.45468753576278687,0.12777777016162872,0.9994792342185974,0.9722222238779068],"occluded":false},{"timestamp":13.26,"box":[0.46562501788139343,0.13333332538604736,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":13.52,"box":[0.4515625238418579,0.11481481045484543,0.9994792342185974,0.972222201526165],"occluded":false},{"timestamp":13.780000000000001,"box":[0.3760416805744171,0.07129629701375961,0.999479204416275,0.9722222313284874],"occluded":false},{"timestamp":14.040000000000001,"box":[0.35104167461395264,0.06481481343507767,0.9994792342185974,0.9722222164273262],"occluded":false},{"timestamp":14.3,"box":[0.47812503576278687,0.05925925821065903,0.9994792342185974,0.9722222313284874],"occluded":false},{"timestamp":14.56,"box":[0.48229169845581055,0.0824074074625969,0.9994792342185974,0.972222201526165],"occluded":false},{"timestamp":14.82,"box":[0.483854204416275,0.10277777910232544,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":15.08,"box":[0.4755208492279053,0.13055555522441864,0.9994792342185974,0.9722221940755844],"occluded":false},{"timestamp":15.34,"box":[0.452604204416275,0.14444443583488464,0.999479204416275,0.9722221791744232],"occluded":false},{"timestamp":15.6,"box":[0.4583333432674408,0.15833333134651184,0.999479204416275,0.9722221791744232],"occluded":false},{"timestamp":15.860000000000001,"box":[0.45468753576278687,0.1527777761220932,0.9994792342185974,0.9712962955236435],"occluded":false},{"timestamp":16.12,"box":[0.46562501788139343,0.16296295821666718,0.999479204416275,0.9712962955236435],"occluded":false},{"timestamp":16.38,"box":[0.46562501788139343,0.15833333134651184,0.999479204416275,0.9712962806224823],"occluded":false},{"timestamp":16.64,"box":[0.45885419845581055,0.15833333134651184,0.9994792342185974,0.9722221791744232],"occluded":false},{"timestamp":16.9,"box":[0.4645833671092987,0.16018518805503845,0.999479204416275,0.9712962806224823],"occluded":false},{"timestamp":17.16,"box":[0.46927085518836975,0.16388888657093048,0.999479204416275,0.9722222238779068],"occluded":false},{"timestamp":17.42,"box":[0.4755208492279053,0.15648147463798523,0.9994792342185974,0.9722221791744232],"occluded":false},{"timestamp":17.68,"box":[0.47864586114883423,0.14999999105930328,0.9994792342185974,0.9722222238779068],"occluded":false},{"timestamp":17.94,"box":[0.47968751192092896,0.15555554628372192,0.9994792342185974,0.9712962508201599],"occluded":false},{"timestamp":18.2,"box":[0.46250003576278687,0.12314814329147339,0.9994792342185974,0.9712962508201599],"occluded":false},{"timestamp":18.46,"box":[0.3348958492279053,0.03703703731298447,0.9994792342185974,0.9712962731719017],"occluded":false},{"timestamp":18.72,"box":[0.4114583432674408,0.031481482088565826,0.999479204416275,0.9722221866250038],"occluded":false},{"timestamp":18.98,"box":[0.43958336114883423,0.03333333134651184,0.9994792342185974,0.9712962806224823],"occluded":false},{"timestamp":19.240000000000002,"box":[0.43177086114883423,0.09259258955717087,0.9994792342185974,0.9712963029742241],"occluded":false},{"timestamp":19.5,"box":[0.4359375238418579,0.14074073731899261,0.9994792342185974,0.9722221940755844],"occluded":false},{"timestamp":19.76,"box":[0.4333333671092987,0.18425925076007843,0.999479204416275,0.9722222238779068],"occluded":false},{"timestamp":20.02,"box":[0.4442708492279053,0.18796296417713165,0.9994792342185974,0.9722222238779068],"occluded":false},{"timestamp":20.28,"box":[0.4567708671092987,0.17314814031124115,0.999479204416275,0.9722221940755844],"occluded":false}],"mean_score":0.9726027250289917},{"noun_phrase":"dog","object_id":0,"presentation_id":3,"observations":[{"timestamp":4.94,"box":[0.35625001788139343,0.7574073672294617,0.38750001788139343,0.7935184761881828],"occluded":false},{"timestamp":5.2,"box":[0.36250001192092896,0.7527777552604675,0.39791668206453323,0.7861110866069794],"occluded":false},{"timestamp":5.46,"box":[0.35364586114883423,0.7407407164573669,0.3942708633840084,0.7796296030282974],"occluded":false},{"timestamp":5.72,"box":[0.35885417461395264,0.7342592477798462,0.39010417461395264,0.7768518403172493],"occluded":false},{"timestamp":5.98,"box":[0.36770835518836975,0.7518518567085266,0.38802085630595684,0.7833333387970924],"occluded":false},{"timestamp":6.5,"box":[0.3541666865348816,0.7277777791023254,0.3817708548158407,0.7750000022351742],"occluded":false},{"timestamp":6.76,"box":[0.3463541865348816,0.7175925970077515,0.37500002048909664,0.7675925977528095],"occluded":false},{"timestamp":7.28,"box":[0.30885419249534607,0.7240740656852722,0.3385416939854622,0.7611111029982567],"occluded":false},{"timestamp":11.700000000000001,"box":[0.3109375238418579,0.7268518209457397,0.35572919249534607,0.7759258933365345],"occluded":false},{"timestamp":11.96,"box":[0.30416667461395264,0.7268518209457397,0.3182291751727462,0.7694444134831429],"occluded":false},{"timestamp":12.22,"box":[0.3062500059604645,0.7157407402992249,0.31822917331010103,0.7546296268701553],"occluded":false},{"timestamp":12.48,"box":[0.3192708492279053,0.7175925970077515,0.33385418355464935,0.7555555589497089],"occluded":false},{"timestamp":12.74,"box":[0.3333333432674408,0.7129629254341125,0.3458333443850279,0.7453703321516514],"occluded":false},{"timestamp":13.0,"box":[0.3447916805744171,0.6990740895271301,0.35625001415610313,0.7388889044523239],"occluded":false},{"timestamp":13.52,"box":[0.3270833492279053,0.7064814567565918,0.3354166829958558,0.7462962716817856],"occluded":false}],"mean_score":0.7928802371025085},{"noun_phrase":"woman","object_id":0,"presentation_id":4,"observations":[{"timestamp":0.0,"box":[0.0,0.23240740597248077,0.25,0.9712962657213211],"occluded":false},{"timestamp":0.26,"box":[0.0,0.1953703612089157,0.27812501788139343,0.9722221940755844],"occluded":false},{"timestamp":0.52,"box":[0.0,0.2203703671693802,0.3786458671092987,0.9749999791383743],"occluded":false},{"timestamp":0.78,"box":[0.0,0.25833332538604736,0.42656251788139343,0.9712962508201599],"occluded":false},{"timestamp":1.04,"box":[0.04270833730697632,0.33981481194496155,0.41822919249534607,0.9712962806224823],"occluded":false},{"timestamp":1.3,"box":[0.0,0.3712962865829468,0.3942708671092987,0.9731481075286865],"occluded":false},{"timestamp":1.56,"box":[0.0,0.364814817905426,0.37708336114883423,0.9722222089767456],"occluded":false},{"timestamp":1.82,"box":[0.0,0.3083333373069763,0.3656250238418579,0.9731481671333313],"occluded":false},{"timestamp":2.08,"box":[0.0,0.2657407224178314,0.36666667461395264,0.9731481373310089],"occluded":false},{"timestamp":2.34,"box":[0.0,0.22407406568527222,0.9395833611488342,0.9722222089767456],"occluded":false},{"timestamp":2.6,"box":[0.0,0.14814814925193787,0.4401041865348816,0.9722221791744232],"occluded":false},{"timestamp":2.86,"box":[0.0,0.18425925076007843,0.4479166865348816,0.9740740805864334],"occluded":false},{"timestamp":3.12,"box":[0.00729166716337204,0.2527777850627899,0.45364584773778915,0.972222238779068],"occluded":false},{"timestamp":3.38,"box":[0.0,0.2925925850868225,0.4322916865348816,0.9731481075286865],"occluded":false},{"timestamp":3.64,"box":[0.0,0.3157407343387604,0.40781253576278687,0.9731481373310089],"occluded":false},{"timestamp":3.9,"box":[0.0,0.3166666626930237,0.3661458492279053,0.9740740656852722],"occluded":false},{"timestamp":4.16,"box":[0.0,0.31296294927597046,0.35208335518836975,0.9722222089767456],"occluded":false},{"timestamp":4.42,"box":[0.0,0.3166666626930237,0.3604166805744171,0.9731481075286865],"occluded":false},{"timestamp":4.68,"box":[0.0,0.2898148000240326,0.3411458432674408,0.9722221791744232],"occluded":false},{"timestamp":4.94,"box":[0.0,0.2787036895751953,0.35364586114883423,0.9722222089767456],"occluded":false},{"timestamp":5.2,"box":[0.0,0.26851850748062134,0.3682291805744171,0.9722222089767456],"occluded":false},{"timestamp":5.46,"box":[0.0,0.2666666507720947,0.39635419845581055,0.9722222089767456],"occluded":false},{"timestamp":5.72,"box":[0.0,0.29629629850387573,0.3916666805744171,0.9740740656852722],"occluded":false},{"timestamp":5.98,"box":[0.0,0.3018518388271332,0.41822919249534607,0.9712962806224823],"occluded":false},{"timestamp":6.24,"box":[0.0,0.34166666865348816,0.43020835518836975,0.9712962806224823],"occluded":false},{"timestamp":6.5,"box":[0.0,0.32407405972480774,0.37968751788139343,0.9722221791744232],"occluded":false},{"timestamp":6.76,"box":[0.0,0.031481482088565826,0.3526041805744171,0.9750000014901161],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.0,0.031481482088565826,0.48072919249534607,0.9722221866250038],"occluded":false},{"timestamp":7.28,"box":[0.0,0.17129629850387573,0.3802083432674408,0.9712963104248047],"occluded":false},{"timestamp":7.54,"box":[0.0,0.1675925850868225,0.41354167461395264,0.9731481075286865],"occluded":false},{"timestamp":7.8,"box":[0.0,0.1675925850868225,0.40572917461395264,0.9722222089767456],"occluded":false},{"timestamp":8.06,"box":[0.0,0.17129629850387573,0.37812501192092896,0.9722222089767456],"occluded":false},{"timestamp":8.32,"box":[0.0,0.25462961196899414,0.3802083432674408,0.9722222089767456],"occluded":false},{"timestamp":8.58,"box":[0.0,0.2648148238658905,0.39270836114883423,0.9990740716457367],"occluded":false},{"timestamp":8.84,"box":[0.0,0.24629628658294678,0.3838541805744171,0.9990740418434143],"occluded":false},{"timestamp":9.1,"box":[0.006770833861082792,0.21574074029922485,0.3802083577029407,0.9722222089767456],"occluded":false},{"timestamp":9.36,"box":[0.0,0.21481481194496155,0.3802083432674408,0.9722221791744232],"occluded":false},{"timestamp":9.620000000000001,"box":[0.0,0.21666666865348816,0.39114585518836975,0.9712962806224823],"occluded":false},{"timestamp":9.88,"box":[0.0,0.18796296417713165,0.39375001192092896,0.9712962657213211],"occluded":false},{"timestamp":10.14,"box":[0.0,0.15555554628372192,0.3776041865348816,0.9990740418434143],"occluded":false},{"timestamp":10.4,"box":[0.0,0.1675925850868225,0.3895833492279053,0.9722222089767456],"occluded":false},{"timestamp":10.66,"box":[0.0,0.17037037014961243,0.3958333432674408,0.9722221791744232],"occluded":false},{"timestamp":10.92,"box":[0.0,0.1675925850868225,0.3734375238418579,0.9712962508201599],"occluded":false},{"timestamp":11.18,"box":[0.0,0.19629628956317902,0.34427085518836975,0.9722222238779068],"occluded":false},{"timestamp":11.44,"box":[0.0,0.2083333283662796,0.3265625238418579,0.9722222238779068],"occluded":false},{"timestamp":11.700000000000001,"box":[0.0,0.18425925076007843,0.32604169845581055,0.9712962657213211],"occluded":false},{"timestamp":11.96,"box":[0.0,0.20185184478759766,0.34947919845581055,0.9722222089767456],"occluded":false},{"timestamp":12.22,"box":[0.0,0.17037037014961243,0.36770835518836975,0.9712962806224823],"occluded":false},{"timestamp":12.48,"box":[0.0,0.05648148059844971,0.4166666865348816,0.9722222089767456],"occluded":false},{"timestamp":12.74,"box":[0.0,0.16296295821666718,0.4218750298023224,0.9722221940755844],"occluded":false},{"timestamp":13.0,"box":[0.0,0.1537037044763565,0.4208333492279053,0.9722222238779068],"occluded":false},{"timestamp":13.26,"box":[0.0,0.18611110746860504,0.43177086114883423,0.9722222238779068],"occluded":false},{"timestamp":13.52,"box":[0.0,0.2074074000120163,0.41875001788139343,0.9712962955236435],"occluded":false},{"timestamp":13.780000000000001,"box":[0.0,0.20462962985038757,0.43958336114883423,0.9712962806224823],"occluded":false},{"timestamp":14.040000000000001,"box":[0.0,0.20277777314186096,0.4208333492279053,0.9712962806224823],"occluded":false},{"timestamp":14.3,"box":[0.0,0.21388888359069824,0.3864583671092987,0.9722222089767456],"occluded":false},{"timestamp":14.56,"box":[0.0,0.21851851046085358,0.3854166865348816,0.9712962657213211],"occluded":false},{"timestamp":14.82,"box":[0.0,0.182407408952713,0.39791667461395264,0.972222238779068],"occluded":false},{"timestamp":15.08,"box":[0.0,0.23148147761821747,0.3958333432674408,0.9722221940755844],"occluded":false},{"timestamp":15.34,"box":[0.0,0.2370370328426361,0.41822919249534607,0.9722221791744232],"occluded":false},{"timestamp":15.6,"box":[0.0,0.24444444477558136,0.43281251192092896,0.9722222238779068],"occluded":false},{"timestamp":15.860000000000001,"box":[0.0,0.2083333283662796,0.42500001192092896,0.9722222238779068],"occluded":false},{"timestamp":16.12,"box":[0.0,0.17499999701976776,0.4463541805744171,0.9722221940755844],"occluded":false},{"timestamp":16.38,"box":[0.0,0.17685185372829437,0.44583335518836975,0.9722221940755844],"occluded":false},{"timestamp":16.64,"box":[0.0,0.12314814329147339,0.44218751788139343,0.9722222089767456],"occluded":false},{"timestamp":16.9,"box":[0.0,0.17407406866550446,0.643750011920929,0.9712962657213211],"occluded":false},{"timestamp":17.16,"box":[0.0031250002793967724,0.2527777850627899,0.42916669277474284,0.972222238779068],"occluded":false},{"timestamp":17.42,"box":[0.0,0.25,0.4192708432674408,0.9712963104248047],"occluded":false},{"timestamp":17.68,"box":[0.0,0.2370370328426361,0.3994791805744171,0.9722221791744232],"occluded":false},{"timestamp":17.94,"box":[0.0,0.1935185194015503,0.36406251788139343,0.9712963104248047],"occluded":false},{"timestamp":18.2,"box":[0.0,0.05092592537403107,0.3906250298023224,0.9722222238779068],"occluded":false},{"timestamp":18.46,"box":[0.0,0.1259259283542633,0.40833336114883423,0.972222238779068],"occluded":false},{"timestamp":18.72,"box":[0.0,0.13333332538604736,0.43802085518836975,0.9712962508201599],"occluded":false},{"timestamp":18.98,"box":[0.0,0.12314814329147339,0.44687503576278687,0.9722222089767456],"occluded":false},{"timestamp":19.240000000000002,"box":[0.0,0.13425925374031067,0.46302086114883423,0.9722221791744232],"occluded":false},{"timestamp":19.5,"box":[0.0,0.1379629671573639,0.4515625238418579,0.972222238779068],"occluded":false},{"timestamp":19.76,"box":[0.0,0.138888880610466,0.45364585518836975,0.9722221940755844],"occluded":false},{"timestamp":20.02,"box":[0.0,0.1787036955356598,0.4609375298023224,0.9722221791744232],"occluded":false},{"timestamp":20.28,"box":[0.0,0.2231481373310089,0.47083336114883423,0.9722221791744232],"occluded":false}],"mean_score":0.9375000596046448}],"dropped":{},"fates":[{"noun_phrase":"dog","object_id":0,"frames":15,"mean_score":0.7928802371025085,"mean_area":0.0009802919477128525,"outcome":"kept","presentation_id":3,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"man with a beard","object_id":0,"frames":79,"mean_score":0.9726027250289917,"mean_area":0.465743725024874,"outcome":"kept","presentation_id":2,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"woman","object_id":0,"frames":79,"mean_score":0.9375000596046448,"mean_area":0.3110538977392104,"outcome":"kept","presentation_id":4,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"woman","object_id":1,"frames":13,"mean_score":0.6683937907218933,"mean_area":0.0022866958382642025,"outcome":"kept","presentation_id":1,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null}],"gates":{"min_track_frames":3,"min_mean_score":0.5,"merge_iou":0.8,"max_tracklets":12}}},"render":{"inputs":{"tracklets":4,"as_video":true,"fps":1.0,"max_frames":30},"outputs":{"media":"/home/zzk/data/video_bbox/vlm_gemini3.5_lite/renders/--cif7Nv6BI_chunk_015_scene_004_choices.mp4","media_bytes":7185135,"frames":[{"index":1,"timestamp":0.0,"file":"--cif7Nv6BI_chunk_015_scene_004_choices.mp4"},{"index":2,"timestamp":1.04,"file":"--cif7Nv6BI_chunk_015_scene_004_choices.mp4"},{"index":3,"timestamp":2.08,"file":"--cif7Nv6BI_chunk_015_scene_004_choices.mp4"},{"index":4,"timestamp":3.12,"file":"--cif7Nv6BI_chunk_015_scene_004_choices.mp4"},{"index":5,"timestamp":4.16,"file":"--cif7Nv6BI_chunk_015_scene_004_choices.mp4"},{"index":6,"timestamp":5.2,"file":"--cif7Nv6BI_chunk_015_scene_004_choices.mp4"},{"index":7,"timestamp":5.98,"file":"--cif7Nv6BI_chunk_015_scene_004_choices.mp4"},{"index":8,"timestamp":7.0200000000000005,"file":"--cif7Nv6BI_chunk_015_scene_004_choices.mp4"},{"index":9,"timestamp":8.06,"file":"--cif7Nv6BI_chunk_015_scene_004_choices.mp4"},{"index":10,"timestamp":9.1,"file":"--cif7Nv6BI_chunk_015_scene_004_choices.mp4"},{"index":11,"timestamp":10.14,"file":"--cif7Nv6BI_chunk_015_scene_004_choices.mp4"},{"index":12,"timestamp":11.18,"file":"--cif7Nv6BI_chunk_015_scene_004_choices.mp4"},{"index":13,"timestamp":12.22,"file":"--cif7Nv6BI_chunk_015_scene_004_choices.mp4"},{"index":14,"timestamp":13.26,"file":"--cif7Nv6BI_chunk_015_scene_004_choices.mp4"},{"index":15,"timestamp":14.3,"file":"--cif7Nv6BI_chunk_015_scene_004_choices.mp4"},{"index":16,"timestamp":15.08,"file":"--cif7Nv6BI_chunk_015_scene_004_choices.mp4"},{"index":17,"timestamp":16.12,"file":"--cif7Nv6BI_chunk_015_scene_004_choices.mp4"},{"index":18,"timestamp":17.16,"file":"--cif7Nv6BI_chunk_015_scene_004_choices.mp4"},{"index":19,"timestamp":18.2,"file":"--cif7Nv6BI_chunk_015_scene_004_choices.mp4"},{"index":20,"timestamp":19.240000000000002,"file":"--cif7Nv6BI_chunk_015_scene_004_choices.mp4"},{"index":21,"timestamp":20.28,"file":"--cif7Nv6BI_chunk_015_scene_004_choices.mp4"}]}},"select":{"inputs":{"model":"gemini-3.5-flash-lite","temperature":0.0,"as_video":true,"media_bytes":7185135,"media":["--cif7Nv6BI_chunk_015_scene_004_choices.mp4"],"choice_frames":21,"system_prompt":"You are labelling video data for referring multi-object tracking.\n\nYou will see a numbered sequence of frames from one clip, in time order. Each frame has its own number and timestamp printed in the top-left corner. Candidate objects have been detected for you and drawn on the frames as coloured boxes, each labelled with a number. The same object keeps the same number and the same colour on every frame. Below each frame is a strip of enlarged crops of those same numbered boxes, so you can identify a small or distant object.\n\nFor each description you are given, answer two questions:\n\n1. MATCHING_IDS — which of the numbered objects the description refers to. List every number that matches, or an EMPTY LIST if none of them do. An empty list is a correct and expected answer: some descriptions are deliberately about something that is not in this clip. Do not pick a number just to have picked one, and do not pick a number for an object that merely resembles the description.\n\n2. START_FRAME and END_FRAME — the frame numbers between which the described state actually holds. If the description is \"the car that's turning left\", the interval is the frames in which that car is turning, NOT every frame in which the car is visible. Use the frame numbers printed on the frames. START_FRAME must not be greater than END_FRAME. When MATCHING_IDS is empty, give the full range (the first and last frame numbers); it is ignored.\n\n3. CONFIDENCE — how sure you are of the MATCHING_IDS answer, from 0.0 to 1.0. Be honest: a low number costs the example nothing, and an overconfident wrong answer becomes a wrong label.\n\nRules:\n- Only use numbers that are actually drawn on the frames.\n- Never output coordinates, box values, or pixel positions.\n- Judge motion by comparing the object's position across frames, not by guessing from one frame.\n","user_prompt":"The video has 21 frames, numbered 1 to 21. Each frame is one moment of the clip and has its number burned into the top left corner.\n\nAnswer for each of the following descriptions, in the order given:\n1. the man with a long beard talking to the camera\n2. the woman standing beside the man\n3. the dog sitting on the ground","questions":["the man with a long beard talking to the camera","the woman standing beside the man","the dog sitting on the ground"]},"outputs":{"raw_response":"{\n  \"answers\": [\n    {\n      \"description_index\": 1,\n      \"matching_ids\": [\n        2\n      ],\n      \"start_frame\": 1,\n      \"end_frame\": 21,\n      \"confidence\": 1.0\n    },\n    {\n      \"description_index\": 2,\n      \"matching_ids\": [\n        4\n      ],\n      \"start_frame\": 1,\n      \"end_frame\": 21,\n      \"confidence\": 1.0\n    },\n    {\n      \"description_index\": 3,\n      \"matching_ids\": [],\n      \"start_frame\": 1,\n      \"end_frame\": 21,\n      \"confidence\": 1.0\n    }\n  ]\n}","raw_answers":[{"description_index":1,"matching_ids":[2],"start_frame":1,"end_frame":21,"confidence":1.0},{"description_index":2,"matching_ids":[4],"start_frame":1,"end_frame":21,"confidence":1.0},{"description_index":3,"matching_ids":[],"start_frame":1,"end_frame":21,"confidence":1.0}],"answers":[{"expression":"the man with a long beard talking to the camera","matching_presentation_ids":[2],"start_choice_frame":1,"end_choice_frame":21,"confidence":1.0,"self_reported_confidence":1.0},{"expression":"the woman standing beside the man","matching_presentation_ids":[4],"start_choice_frame":1,"end_choice_frame":21,"confidence":1.0,"self_reported_confidence":1.0},{"expression":"the dog sitting on the ground","matching_presentation_ids":[],"start_choice_frame":1,"end_choice_frame":21,"confidence":1.0,"self_reported_confidence":1.0}]}},"verify":{"inputs":{"expressions":3,"tracklets":4,"selections":3,"gates":"interval, count, unknown ids, confidence, then geometry (motion)"},"outputs":{"kept":[{"expression":"the man with a long beard talking to the camera","kind":"appearance","is_negative":false,"tracks":[{"timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88,10.14,10.4,10.66,10.92,11.18,11.44,11.700000000000001,11.96,12.22,12.48,12.74,13.0,13.26,13.52,13.780000000000001,14.040000000000001,14.3,14.56,14.82,15.08,15.34,15.6,15.860000000000001,16.12,16.38,16.64,16.9,17.16,17.42,17.68,17.94,18.2,18.46,18.72,18.98,19.240000000000002,19.5,19.76,20.02,20.28],"boxes":[[0.5562500357627869,0.08148147910833359,0.9994792342185974,0.9703703746199608],[0.3437500298023224,0.20462962985038757,0.9708333909511566,0.9712962806224823],[0.3687500059604645,0.1787036955356598,0.999479204416275,0.9722221791744232],[0.4453125298023224,0.06759259104728699,0.999479204416275,0.9712962806224823],[0.4515625238418579,0.10740740597248077,0.9885417222976685,0.9722222238779068],[0.4401041865348816,0.1101851835846901,0.9708333611488342,0.9722221866250038],[0.4609375298023224,0.04537037014961243,0.999479204416275,0.9722221791744232],[0.45729169249534607,0.13518518209457397,0.9895834028720856,0.9722222089767456],[0.45625001192092896,0.1944444477558136,0.9781250357627869,0.972222238779068],[0.45468753576278687,0.10833333432674408,0.97552090883255,0.9722221940755844],[0.4505208432674408,0.09259258955717087,0.9708333909511566,0.972222201526165],[0.4437500238418579,0.13611111044883728,0.9729167222976685,0.9722221791744232],[0.44166669249534607,0.1537037044763565,0.9828125536441803,0.9722222238779068],[0.45625001192092896,0.12314814329147339,0.9994792342185974,0.9722222089767456],[0.45781251788139343,0.12870369851589203,0.999479204416275,0.9722221940755844],[0.4593750238418579,0.12870369851589203,0.9994792342185974,0.9722221940755844],[0.4598958492279053,0.15833333134651184,0.9994792342185974,0.9722221791744232],[0.4583333432674408,0.11574073880910873,0.999479204416275,0.9722222313284874],[0.4453125298023224,0.09166666865348816,0.999479204416275,0.9722221791744232],[0.4375000298023224,0.13148148357868195,0.999479204416275,0.9722222238779068],[0.4479166865348816,0.09444444626569748,0.9994792342185974,0.972222201526165],[0.44583335518836975,0.07129629701375961,0.999479204416275,0.9722222313284874],[0.45104169845581055,0.10833333432674408,0.9994792342185974,0.9722221940755844],[0.43281251192092896,0.13240741193294525,0.9937500357627869,0.9722221940755844],[0.4322916865348816,0.13333332538604736,0.9994792342185974,0.9722222089767456],[0.43437501788139343,0.1537037044763565,0.999479204416275,0.9722222238779068],[0.42760419845581055,0.12870369851589203,0.9994792342185974,0.9722221940755844],[0.44583335518836975,0.14722222089767456,0.999479204416275,0.9722222089767456],[0.4541666805744171,0.13611111044883728,0.999479204416275,0.9722221791744232],[0.45625001192092896,0.09907407313585281,0.9994792342185974,0.9712962582707405],[0.4593750238418579,0.11296296119689941,0.9994792342185974,0.9712962508201599],[0.4598958492279053,0.08981481194496155,0.9994792342185974,0.9712962806224823],[0.4661458432674408,0.13703703880310059,0.999479204416275,0.9722222089767456],[0.4619791805744171,0.09351851791143417,0.999479204416275,0.9712962731719017],[0.4598958492279053,0.09351851791143417,0.9994792342185974,0.9722222313284874],[0.4557291865348816,0.08703703433275223,0.9994792342185974,0.9722222164273262],[0.45104169845581055,0.07870370149612427,0.9994792342185974,0.9722222089767456],[0.38333335518836975,0.14999999105930328,0.999479204416275,0.9722222238779068],[0.37552085518836975,0.15648147463798523,0.999479204416275,0.9722221791744232],[0.4114583432674408,0.138888880610466,0.9968750178813934,0.9722221940755844],[0.4375000298023224,0.16018518805503845,0.999479204416275,0.972222238779068],[0.4489583671092987,0.20000000298023224,0.999479204416275,0.9722222238779068],[0.45468753576278687,0.14814814925193787,0.9994792342185974,0.9722221791744232],[0.43697917461395264,0.1379629671573639,0.9994791746139526,0.972222238779068],[0.4520833492279053,0.14166666567325592,0.9994792342185974,0.9722222238779068],[0.4479166865348816,0.14722222089767456,0.9994792342185974,0.9722222089767456],[0.44583335518836975,0.16944444179534912,0.999479204416275,0.9722222089767456],[0.46406251192092896,0.15648147463798523,0.9994792342185974,0.9722221791744232],[0.4598958492279053,0.13703703880310059,0.9994792342185974,0.9722222089767456],[0.4505208432674408,0.14074073731899261,0.999479204416275,0.9722221940755844],[0.45468753576278687,0.12777777016162872,0.9994792342185974,0.9722222238779068],[0.46562501788139343,0.13333332538604736,0.999479204416275,0.9722222089767456],[0.4515625238418579,0.11481481045484543,0.9994792342185974,0.972222201526165],[0.3760416805744171,0.07129629701375961,0.999479204416275,0.9722222313284874],[0.35104167461395264,0.06481481343507767,0.9994792342185974,0.9722222164273262],[0.47812503576278687,0.05925925821065903,0.9994792342185974,0.9722222313284874],[0.48229169845581055,0.0824074074625969,0.9994792342185974,0.972222201526165],[0.483854204416275,0.10277777910232544,0.999479204416275,0.9722222089767456],[0.4755208492279053,0.13055555522441864,0.9994792342185974,0.9722221940755844],[0.452604204416275,0.14444443583488464,0.999479204416275,0.9722221791744232],[0.4583333432674408,0.15833333134651184,0.999479204416275,0.9722221791744232],[0.45468753576278687,0.1527777761220932,0.9994792342185974,0.9712962955236435],[0.46562501788139343,0.16296295821666718,0.999479204416275,0.9712962955236435],[0.46562501788139343,0.15833333134651184,0.999479204416275,0.9712962806224823],[0.45885419845581055,0.15833333134651184,0.9994792342185974,0.9722221791744232],[0.4645833671092987,0.16018518805503845,0.999479204416275,0.9712962806224823],[0.46927085518836975,0.16388888657093048,0.999479204416275,0.9722222238779068],[0.4755208492279053,0.15648147463798523,0.9994792342185974,0.9722221791744232],[0.47864586114883423,0.14999999105930328,0.9994792342185974,0.9722222238779068],[0.47968751192092896,0.15555554628372192,0.9994792342185974,0.9712962508201599],[0.46250003576278687,0.12314814329147339,0.9994792342185974,0.9712962508201599],[0.3348958492279053,0.03703703731298447,0.9994792342185974,0.9712962731719017],[0.4114583432674408,0.031481482088565826,0.999479204416275,0.9722221866250038],[0.43958336114883423,0.03333333134651184,0.9994792342185974,0.9712962806224823],[0.43177086114883423,0.09259258955717087,0.9994792342185974,0.9712963029742241],[0.4359375238418579,0.14074073731899261,0.9994792342185974,0.9722221940755844],[0.4333333671092987,0.18425925076007843,0.999479204416275,0.9722222238779068],[0.4442708492279053,0.18796296417713165,0.9994792342185974,0.9722222238779068],[0.4567708671092987,0.17314814031124115,0.999479204416275,0.9722221940755844]]}]},{"expression":"the woman standing beside the man","kind":"spatial","is_negative":false,"tracks":[{"timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88,10.14,10.4,10.66,10.92,11.18,11.44,11.700000000000001,11.96,12.22,12.48,12.74,13.0,13.26,13.52,13.780000000000001,14.040000000000001,14.3,14.56,14.82,15.08,15.34,15.6,15.860000000000001,16.12,16.38,16.64,16.9,17.16,17.42,17.68,17.94,18.2,18.46,18.72,18.98,19.240000000000002,19.5,19.76,20.02,20.28],"boxes":[[0.0,0.23240740597248077,0.25,0.9712962657213211],[0.0,0.1953703612089157,0.27812501788139343,0.9722221940755844],[0.0,0.2203703671693802,0.3786458671092987,0.9749999791383743],[0.0,0.25833332538604736,0.42656251788139343,0.9712962508201599],[0.04270833730697632,0.33981481194496155,0.41822919249534607,0.9712962806224823],[0.0,0.3712962865829468,0.3942708671092987,0.9731481075286865],[0.0,0.364814817905426,0.37708336114883423,0.9722222089767456],[0.0,0.3083333373069763,0.3656250238418579,0.9731481671333313],[0.0,0.2657407224178314,0.36666667461395264,0.9731481373310089],[0.0,0.22407406568527222,0.9395833611488342,0.9722222089767456],[0.0,0.14814814925193787,0.4401041865348816,0.9722221791744232],[0.0,0.18425925076007843,0.4479166865348816,0.9740740805864334],[0.00729166716337204,0.2527777850627899,0.45364584773778915,0.972222238779068],[0.0,0.2925925850868225,0.4322916865348816,0.9731481075286865],[0.0,0.3157407343387604,0.40781253576278687,0.9731481373310089],[0.0,0.3166666626930237,0.3661458492279053,0.9740740656852722],[0.0,0.31296294927597046,0.35208335518836975,0.9722222089767456],[0.0,0.3166666626930237,0.3604166805744171,0.9731481075286865],[0.0,0.2898148000240326,0.3411458432674408,0.9722221791744232],[0.0,0.2787036895751953,0.35364586114883423,0.9722222089767456],[0.0,0.26851850748062134,0.3682291805744171,0.9722222089767456],[0.0,0.2666666507720947,0.39635419845581055,0.9722222089767456],[0.0,0.29629629850387573,0.3916666805744171,0.9740740656852722],[0.0,0.3018518388271332,0.41822919249534607,0.9712962806224823],[0.0,0.34166666865348816,0.43020835518836975,0.9712962806224823],[0.0,0.32407405972480774,0.37968751788139343,0.9722221791744232],[0.0,0.031481482088565826,0.3526041805744171,0.9750000014901161],[0.0,0.031481482088565826,0.48072919249534607,0.9722221866250038],[0.0,0.17129629850387573,0.3802083432674408,0.9712963104248047],[0.0,0.1675925850868225,0.41354167461395264,0.9731481075286865],[0.0,0.1675925850868225,0.40572917461395264,0.9722222089767456],[0.0,0.17129629850387573,0.37812501192092896,0.9722222089767456],[0.0,0.25462961196899414,0.3802083432674408,0.9722222089767456],[0.0,0.2648148238658905,0.39270836114883423,0.9990740716457367],[0.0,0.24629628658294678,0.3838541805744171,0.9990740418434143],[0.006770833861082792,0.21574074029922485,0.3802083577029407,0.9722222089767456],[0.0,0.21481481194496155,0.3802083432674408,0.9722221791744232],[0.0,0.21666666865348816,0.39114585518836975,0.9712962806224823],[0.0,0.18796296417713165,0.39375001192092896,0.9712962657213211],[0.0,0.15555554628372192,0.3776041865348816,0.9990740418434143],[0.0,0.1675925850868225,0.3895833492279053,0.9722222089767456],[0.0,0.17037037014961243,0.3958333432674408,0.9722221791744232],[0.0,0.1675925850868225,0.3734375238418579,0.9712962508201599],[0.0,0.19629628956317902,0.34427085518836975,0.9722222238779068],[0.0,0.2083333283662796,0.3265625238418579,0.9722222238779068],[0.0,0.18425925076007843,0.32604169845581055,0.9712962657213211],[0.0,0.20185184478759766,0.34947919845581055,0.9722222089767456],[0.0,0.17037037014961243,0.36770835518836975,0.9712962806224823],[0.0,0.05648148059844971,0.4166666865348816,0.9722222089767456],[0.0,0.16296295821666718,0.4218750298023224,0.9722221940755844],[0.0,0.1537037044763565,0.4208333492279053,0.9722222238779068],[0.0,0.18611110746860504,0.43177086114883423,0.9722222238779068],[0.0,0.2074074000120163,0.41875001788139343,0.9712962955236435],[0.0,0.20462962985038757,0.43958336114883423,0.9712962806224823],[0.0,0.20277777314186096,0.4208333492279053,0.9712962806224823],[0.0,0.21388888359069824,0.3864583671092987,0.9722222089767456],[0.0,0.21851851046085358,0.3854166865348816,0.9712962657213211],[0.0,0.182407408952713,0.39791667461395264,0.972222238779068],[0.0,0.23148147761821747,0.3958333432674408,0.9722221940755844],[0.0,0.2370370328426361,0.41822919249534607,0.9722221791744232],[0.0,0.24444444477558136,0.43281251192092896,0.9722222238779068],[0.0,0.2083333283662796,0.42500001192092896,0.9722222238779068],[0.0,0.17499999701976776,0.4463541805744171,0.9722221940755844],[0.0,0.17685185372829437,0.44583335518836975,0.9722221940755844],[0.0,0.12314814329147339,0.44218751788139343,0.9722222089767456],[0.0,0.17407406866550446,0.643750011920929,0.9712962657213211],[0.0031250002793967724,0.2527777850627899,0.42916669277474284,0.972222238779068],[0.0,0.25,0.4192708432674408,0.9712963104248047],[0.0,0.2370370328426361,0.3994791805744171,0.9722221791744232],[0.0,0.1935185194015503,0.36406251788139343,0.9712963104248047],[0.0,0.05092592537403107,0.3906250298023224,0.9722222238779068],[0.0,0.1259259283542633,0.40833336114883423,0.972222238779068],[0.0,0.13333332538604736,0.43802085518836975,0.9712962508201599],[0.0,0.12314814329147339,0.44687503576278687,0.9722222089767456],[0.0,0.13425925374031067,0.46302086114883423,0.9722221791744232],[0.0,0.1379629671573639,0.4515625238418579,0.972222238779068],[0.0,0.138888880610466,0.45364585518836975,0.9722221940755844],[0.0,0.1787036955356598,0.4609375298023224,0.9722221791744232],[0.0,0.2231481373310089,0.47083336114883423,0.9722221791744232]]}]},{"expression":"the dog sitting on the ground","kind":"appearance","is_negative":true,"tracks":[]}],"rejected":{},"rejection_counts":{},"row":{"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_015_scene_004.mp4","scene_key":"--cif7Nv6BI_chunk_015_scene_004","category":"test_vehicles","height":1080,"width":1920,"fps":50.0,"num_frames":1024,"duration":20.48,"rmot_expressions":[{"expression":"the man with a long beard talking to the camera","kind":"appearance","is_negative":false,"tracks":[{"timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88,10.14,10.4,10.66,10.92,11.18,11.44,11.700000000000001,11.96,12.22,12.48,12.74,13.0,13.26,13.52,13.780000000000001,14.040000000000001,14.3,14.56,14.82,15.08,15.34,15.6,15.860000000000001,16.12,16.38,16.64,16.9,17.16,17.42,17.68,17.94,18.2,18.46,18.72,18.98,19.240000000000002,19.5,19.76,20.02,20.28],"boxes":[[0.5562500357627869,0.08148147910833359,0.9994792342185974,0.9703703746199608],[0.3437500298023224,0.20462962985038757,0.9708333909511566,0.9712962806224823],[0.3687500059604645,0.1787036955356598,0.999479204416275,0.9722221791744232],[0.4453125298023224,0.06759259104728699,0.999479204416275,0.9712962806224823],[0.4515625238418579,0.10740740597248077,0.9885417222976685,0.9722222238779068],[0.4401041865348816,0.1101851835846901,0.9708333611488342,0.9722221866250038],[0.4609375298023224,0.04537037014961243,0.999479204416275,0.9722221791744232],[0.45729169249534607,0.13518518209457397,0.9895834028720856,0.9722222089767456],[0.45625001192092896,0.1944444477558136,0.9781250357627869,0.972222238779068],[0.45468753576278687,0.10833333432674408,0.97552090883255,0.9722221940755844],[0.4505208432674408,0.09259258955717087,0.9708333909511566,0.972222201526165],[0.4437500238418579,0.13611111044883728,0.9729167222976685,0.9722221791744232],[0.44166669249534607,0.1537037044763565,0.9828125536441803,0.9722222238779068],[0.45625001192092896,0.12314814329147339,0.9994792342185974,0.9722222089767456],[0.45781251788139343,0.12870369851589203,0.999479204416275,0.9722221940755844],[0.4593750238418579,0.12870369851589203,0.9994792342185974,0.9722221940755844],[0.4598958492279053,0.15833333134651184,0.9994792342185974,0.9722221791744232],[0.4583333432674408,0.11574073880910873,0.999479204416275,0.9722222313284874],[0.4453125298023224,0.09166666865348816,0.999479204416275,0.9722221791744232],[0.4375000298023224,0.13148148357868195,0.999479204416275,0.9722222238779068],[0.4479166865348816,0.09444444626569748,0.9994792342185974,0.972222201526165],[0.44583335518836975,0.07129629701375961,0.999479204416275,0.9722222313284874],[0.45104169845581055,0.10833333432674408,0.9994792342185974,0.9722221940755844],[0.43281251192092896,0.13240741193294525,0.9937500357627869,0.9722221940755844],[0.4322916865348816,0.13333332538604736,0.9994792342185974,0.9722222089767456],[0.43437501788139343,0.1537037044763565,0.999479204416275,0.9722222238779068],[0.42760419845581055,0.12870369851589203,0.9994792342185974,0.9722221940755844],[0.44583335518836975,0.14722222089767456,0.999479204416275,0.9722222089767456],[0.4541666805744171,0.13611111044883728,0.999479204416275,0.9722221791744232],[0.45625001192092896,0.09907407313585281,0.9994792342185974,0.9712962582707405],[0.4593750238418579,0.11296296119689941,0.9994792342185974,0.9712962508201599],[0.4598958492279053,0.08981481194496155,0.9994792342185974,0.9712962806224823],[0.4661458432674408,0.13703703880310059,0.999479204416275,0.9722222089767456],[0.4619791805744171,0.09351851791143417,0.999479204416275,0.9712962731719017],[0.4598958492279053,0.09351851791143417,0.9994792342185974,0.9722222313284874],[0.4557291865348816,0.08703703433275223,0.9994792342185974,0.9722222164273262],[0.45104169845581055,0.07870370149612427,0.9994792342185974,0.9722222089767456],[0.38333335518836975,0.14999999105930328,0.999479204416275,0.9722222238779068],[0.37552085518836975,0.15648147463798523,0.999479204416275,0.9722221791744232],[0.4114583432674408,0.138888880610466,0.9968750178813934,0.9722221940755844],[0.4375000298023224,0.16018518805503845,0.999479204416275,0.972222238779068],[0.4489583671092987,0.20000000298023224,0.999479204416275,0.9722222238779068],[0.45468753576278687,0.14814814925193787,0.9994792342185974,0.9722221791744232],[0.43697917461395264,0.1379629671573639,0.9994791746139526,0.972222238779068],[0.4520833492279053,0.14166666567325592,0.9994792342185974,0.9722222238779068],[0.4479166865348816,0.14722222089767456,0.9994792342185974,0.9722222089767456],[0.44583335518836975,0.16944444179534912,0.999479204416275,0.9722222089767456],[0.46406251192092896,0.15648147463798523,0.9994792342185974,0.9722221791744232],[0.4598958492279053,0.13703703880310059,0.9994792342185974,0.9722222089767456],[0.4505208432674408,0.14074073731899261,0.999479204416275,0.9722221940755844],[0.45468753576278687,0.12777777016162872,0.9994792342185974,0.9722222238779068],[0.46562501788139343,0.13333332538604736,0.999479204416275,0.9722222089767456],[0.4515625238418579,0.11481481045484543,0.9994792342185974,0.972222201526165],[0.3760416805744171,0.07129629701375961,0.999479204416275,0.9722222313284874],[0.35104167461395264,0.06481481343507767,0.9994792342185974,0.9722222164273262],[0.47812503576278687,0.05925925821065903,0.9994792342185974,0.9722222313284874],[0.48229169845581055,0.0824074074625969,0.9994792342185974,0.972222201526165],[0.483854204416275,0.10277777910232544,0.999479204416275,0.9722222089767456],[0.4755208492279053,0.13055555522441864,0.9994792342185974,0.9722221940755844],[0.452604204416275,0.14444443583488464,0.999479204416275,0.9722221791744232],[0.4583333432674408,0.15833333134651184,0.999479204416275,0.9722221791744232],[0.45468753576278687,0.1527777761220932,0.9994792342185974,0.9712962955236435],[0.46562501788139343,0.16296295821666718,0.999479204416275,0.9712962955236435],[0.46562501788139343,0.15833333134651184,0.999479204416275,0.9712962806224823],[0.45885419845581055,0.15833333134651184,0.9994792342185974,0.9722221791744232],[0.4645833671092987,0.16018518805503845,0.999479204416275,0.9712962806224823],[0.46927085518836975,0.16388888657093048,0.999479204416275,0.9722222238779068],[0.4755208492279053,0.15648147463798523,0.9994792342185974,0.9722221791744232],[0.47864586114883423,0.14999999105930328,0.9994792342185974,0.9722222238779068],[0.47968751192092896,0.15555554628372192,0.9994792342185974,0.9712962508201599],[0.46250003576278687,0.12314814329147339,0.9994792342185974,0.9712962508201599],[0.3348958492279053,0.03703703731298447,0.9994792342185974,0.9712962731719017],[0.4114583432674408,0.031481482088565826,0.999479204416275,0.9722221866250038],[0.43958336114883423,0.03333333134651184,0.9994792342185974,0.9712962806224823],[0.43177086114883423,0.09259258955717087,0.9994792342185974,0.9712963029742241],[0.4359375238418579,0.14074073731899261,0.9994792342185974,0.9722221940755844],[0.4333333671092987,0.18425925076007843,0.999479204416275,0.9722222238779068],[0.4442708492279053,0.18796296417713165,0.9994792342185974,0.9722222238779068],[0.4567708671092987,0.17314814031124115,0.999479204416275,0.9722221940755844]]}]},{"expression":"the woman standing beside the man","kind":"spatial","is_negative":false,"tracks":[{"timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88,10.14,10.4,10.66,10.92,11.18,11.44,11.700000000000001,11.96,12.22,12.48,12.74,13.0,13.26,13.52,13.780000000000001,14.040000000000001,14.3,14.56,14.82,15.08,15.34,15.6,15.860000000000001,16.12,16.38,16.64,16.9,17.16,17.42,17.68,17.94,18.2,18.46,18.72,18.98,19.240000000000002,19.5,19.76,20.02,20.28],"boxes":[[0.0,0.23240740597248077,0.25,0.9712962657213211],[0.0,0.1953703612089157,0.27812501788139343,0.9722221940755844],[0.0,0.2203703671693802,0.3786458671092987,0.9749999791383743],[0.0,0.25833332538604736,0.42656251788139343,0.9712962508201599],[0.04270833730697632,0.33981481194496155,0.41822919249534607,0.9712962806224823],[0.0,0.3712962865829468,0.3942708671092987,0.9731481075286865],[0.0,0.364814817905426,0.37708336114883423,0.9722222089767456],[0.0,0.3083333373069763,0.3656250238418579,0.9731481671333313],[0.0,0.2657407224178314,0.36666667461395264,0.9731481373310089],[0.0,0.22407406568527222,0.9395833611488342,0.9722222089767456],[0.0,0.14814814925193787,0.4401041865348816,0.9722221791744232],[0.0,0.18425925076007843,0.4479166865348816,0.9740740805864334],[0.00729166716337204,0.2527777850627899,0.45364584773778915,0.972222238779068],[0.0,0.2925925850868225,0.4322916865348816,0.9731481075286865],[0.0,0.3157407343387604,0.40781253576278687,0.9731481373310089],[0.0,0.3166666626930237,0.3661458492279053,0.9740740656852722],[0.0,0.31296294927597046,0.35208335518836975,0.9722222089767456],[0.0,0.3166666626930237,0.3604166805744171,0.9731481075286865],[0.0,0.2898148000240326,0.3411458432674408,0.9722221791744232],[0.0,0.2787036895751953,0.35364586114883423,0.9722222089767456],[0.0,0.26851850748062134,0.3682291805744171,0.9722222089767456],[0.0,0.2666666507720947,0.39635419845581055,0.9722222089767456],[0.0,0.29629629850387573,0.3916666805744171,0.9740740656852722],[0.0,0.3018518388271332,0.41822919249534607,0.9712962806224823],[0.0,0.34166666865348816,0.43020835518836975,0.9712962806224823],[0.0,0.32407405972480774,0.37968751788139343,0.9722221791744232],[0.0,0.031481482088565826,0.3526041805744171,0.9750000014901161],[0.0,0.031481482088565826,0.48072919249534607,0.9722221866250038],[0.0,0.17129629850387573,0.3802083432674408,0.9712963104248047],[0.0,0.1675925850868225,0.41354167461395264,0.9731481075286865],[0.0,0.1675925850868225,0.40572917461395264,0.9722222089767456],[0.0,0.17129629850387573,0.37812501192092896,0.9722222089767456],[0.0,0.25462961196899414,0.3802083432674408,0.9722222089767456],[0.0,0.2648148238658905,0.39270836114883423,0.9990740716457367],[0.0,0.24629628658294678,0.3838541805744171,0.9990740418434143],[0.006770833861082792,0.21574074029922485,0.3802083577029407,0.9722222089767456],[0.0,0.21481481194496155,0.3802083432674408,0.9722221791744232],[0.0,0.21666666865348816,0.39114585518836975,0.9712962806224823],[0.0,0.18796296417713165,0.39375001192092896,0.9712962657213211],[0.0,0.15555554628372192,0.3776041865348816,0.9990740418434143],[0.0,0.1675925850868225,0.3895833492279053,0.9722222089767456],[0.0,0.17037037014961243,0.3958333432674408,0.9722221791744232],[0.0,0.1675925850868225,0.3734375238418579,0.9712962508201599],[0.0,0.19629628956317902,0.34427085518836975,0.9722222238779068],[0.0,0.2083333283662796,0.3265625238418579,0.9722222238779068],[0.0,0.18425925076007843,0.32604169845581055,0.9712962657213211],[0.0,0.20185184478759766,0.34947919845581055,0.9722222089767456],[0.0,0.17037037014961243,0.36770835518836975,0.9712962806224823],[0.0,0.05648148059844971,0.4166666865348816,0.9722222089767456],[0.0,0.16296295821666718,0.4218750298023224,0.9722221940755844],[0.0,0.1537037044763565,0.4208333492279053,0.9722222238779068],[0.0,0.18611110746860504,0.43177086114883423,0.9722222238779068],[0.0,0.2074074000120163,0.41875001788139343,0.9712962955236435],[0.0,0.20462962985038757,0.43958336114883423,0.9712962806224823],[0.0,0.20277777314186096,0.4208333492279053,0.9712962806224823],[0.0,0.21388888359069824,0.3864583671092987,0.9722222089767456],[0.0,0.21851851046085358,0.3854166865348816,0.9712962657213211],[0.0,0.182407408952713,0.39791667461395264,0.972222238779068],[0.0,0.23148147761821747,0.3958333432674408,0.9722221940755844],[0.0,0.2370370328426361,0.41822919249534607,0.9722221791744232],[0.0,0.24444444477558136,0.43281251192092896,0.9722222238779068],[0.0,0.2083333283662796,0.42500001192092896,0.9722222238779068],[0.0,0.17499999701976776,0.4463541805744171,0.9722221940755844],[0.0,0.17685185372829437,0.44583335518836975,0.9722221940755844],[0.0,0.12314814329147339,0.44218751788139343,0.9722222089767456],[0.0,0.17407406866550446,0.643750011920929,0.9712962657213211],[0.0031250002793967724,0.2527777850627899,0.42916669277474284,0.972222238779068],[0.0,0.25,0.4192708432674408,0.9712963104248047],[0.0,0.2370370328426361,0.3994791805744171,0.9722221791744232],[0.0,0.1935185194015503,0.36406251788139343,0.9712963104248047],[0.0,0.05092592537403107,0.3906250298023224,0.9722222238779068],[0.0,0.1259259283542633,0.40833336114883423,0.972222238779068],[0.0,0.13333332538604736,0.43802085518836975,0.9712962508201599],[0.0,0.12314814329147339,0.44687503576278687,0.9722222089767456],[0.0,0.13425925374031067,0.46302086114883423,0.9722221791744232],[0.0,0.1379629671573639,0.4515625238418579,0.972222238779068],[0.0,0.138888880610466,0.45364585518836975,0.9722221940755844],[0.0,0.1787036955356598,0.4609375298023224,0.9722221791744232],[0.0,0.2231481373310089,0.47083336114883423,0.9722221791744232]]}]},{"expression":"the dog sitting on the ground","kind":"appearance","is_negative":true,"tracks":[]}],"rmot_expression_count":3,"rmot_negative_count":1,"rmot_motion_count":0}}}}},{"index":11,"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_012_scene_010.mp4","status":"complete","steps":{"decode":{"inputs":{"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_012_scene_010.mp4","sample_fps":4.0},"outputs":{"width":1920,"height":1080,"source_fps":50.0,"duration_seconds":12.92,"frames_kept":50,"blank_frames_dropped":0,"timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88,10.14,10.4,10.66,10.92,11.18,11.44,11.700000000000001,11.96,12.22,12.48,12.74],"error":""}},"propose":{"inputs":{"model":"gemini-3.5-flash-lite","temperature":0.4,"video_bytes":18615977,"system_prompt":"You are annotating video clips to build training data for referring multi-object tracking: given a video and a natural-language description, a model must output the boxes of every object matching that description, over time.\n\nYour job is to write the descriptions. You do NOT output coordinates, boxes, pixel positions, or any numbers about where things are. A separate detector finds the objects.\n\nFor every clip you produce a list of items. Each item has two texts that must NOT be the same:\n\nEXPRESSION — what a person would say to single out the object or objects. It may describe motion, temporal order, interaction with other objects, and spatial relations. It should read like something a person would actually say.\n  Good: \"the black car that's turning left\"\n  Good: \"the cyclist who overtakes the bus\"\n  Good: \"the two people who walk out of frame on the right\"\n\nNOUN_PHRASE — the same object as a plain, timeless visual category with its visible attributes, and NOTHING else. The detector is prompted with this text and looks at ONE frame at a time, so anything about time, order, or action is useless to it and makes it fail.\n  Allowed: category nouns, colours, materials, clothing, size, type\n  Forbidden: verbs and participles of action (\"turning\", \"walks\", \"is moving\"), temporal words (\"then\", \"starts\", \"after\", \"first\", \"last\"), relative clauses (\"that ...\", \"which ...\", \"who ...\"), ordinals and rank words (\"first\", \"leftmost\", \"nearest\"), and anything over six words.\n  \"the black car that's turning left\" -> \"black car\"\n  \"the cyclist who overtakes the bus\" -> \"cyclist\"\n  \"the two people who walk out of frame on the right\" -> \"person\"\n\nKIND — what makes the expression hard: \"motion\", \"appearance\", \"spatial\", or \"interaction\". Aim for about 35% \"motion\" items: motion is the part a single-frame model cannot already solve, so it is the most valuable, but a list of only motion items is not usable training data either.\n\nEXPECTED_COUNT — how many distinct objects in the clip match the EXPRESSION. Count carefully; this is checked against the detector's answer, and a wrong count discards the item.\n\nIS_NEGATIVE — at most ONE item may be a negative, and only when you also write at least one non-negative item. A negative is an expression that recombines attributes that ARE present in the clip into a whole that is NOT. If the only black car is moving, \"the parked black car\" is a good negative. Its NOUN_PHRASE must still be the category that IS present (\"black car\"), because that is what makes the absence checkable rather than merely unverified. A negative's EXPECTED_COUNT is 0.\n\nRules that discard an item, so follow them exactly:\n- Never write coordinates, box values, pixel positions, or percentages of the frame.\n- Never let a NOUN_PHRASE contain a verb of action, a temporal word, a relative clause, or an ordinal.\n- Never write an EXPRESSION whose object is not visible in the clip, unless it is marked IS_NEGATIVE.\n- Prefer objects that are clearly visible and identifiable over background clutter.\n- Do not describe the camera or the shot; describe objects in the scene.\n","user_prompt":"Watch this clip and write between 1 and 3 items following the rules. Cover different objects rather than restating one object several ways, and include any negative in the same list.\n"},"outputs":{"raw_response":"{\n  \"items\": [\n    {\n      \"expression\": \"the huge boulder acting as a roof for the building\",\n      \"noun_phrase\": \"huge boulder\",\n      \"kind\": \"spatial\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the red door of the stone house\",\n      \"noun_phrase\": \"red door\",\n      \"kind\": \"appearance\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the blue car parked in the village square\",\n      \"noun_phrase\": \"blue car\",\n      \"kind\": \"appearance\",\n      \"expected_count\": 0,\n      \"is_negative\": true\n    }\n  ]\n}","raw_items":[{"expression":"the huge boulder acting as a roof for the building","noun_phrase":"huge boulder","kind":"spatial","expected_count":1,"is_negative":false},{"expression":"the red door of the stone house","noun_phrase":"red door","kind":"appearance","expected_count":1,"is_negative":false},{"expression":"the blue car parked in the village square","noun_phrase":"blue car","kind":"appearance","expected_count":0,"is_negative":true}],"accepted":[{"expression":"the huge boulder acting as a roof for the building","noun_phrase":"huge boulder","kind":"spatial","expected_count":1,"is_negative":false},{"expression":"the red door of the stone house","noun_phrase":"red door","kind":"appearance","expected_count":1,"is_negative":false},{"expression":"the blue car parked in the village square","noun_phrase":"blue car","kind":"appearance","expected_count":0,"is_negative":true}],"gated":[]}},"track":{"inputs":{"tracker":"Sam31MultiplexTracker","model_id":"/mnt/ceph3/ec/jacky/sam3.1","noun_phrases":["blue car","huge boulder","red door"],"expressions_in":3,"frames":50,"sample_fps":4.0,"frame_size":"1920x1080","timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88,10.14,10.4,10.66,10.92,11.18,11.44,11.700000000000001,11.96,12.22,12.48,12.74],"session":"one session over 50 JPEG frames; then per phrase: reset_session, add_prompt(text) on frame 0, propagate_in_video","passes":3},"outputs":{"raw_tracks":[{"noun_phrase":"huge boulder","object_id":0,"observations":[{"timestamp":0.0,"box":[0.6322916746139526,0.7046296000480652,0.7744791805744171,0.9231481105089188],"occluded":false},{"timestamp":0.26,"box":[0.6328125596046448,0.7064814567565918,0.7739583998918533,0.9212962687015533],"occluded":false},{"timestamp":0.52,"box":[0.6244791746139526,0.7111110687255859,0.768750011920929,0.9305555075407028],"occluded":false},{"timestamp":0.78,"box":[0.6177083849906921,0.75,0.7734375596046448,0.9703703671693802],"occluded":false},{"timestamp":1.04,"box":[0.6223958730697632,0.7768518328666687,0.7875000536441803,0.9722221940755844],"occluded":false},{"timestamp":1.3,"box":[0.6223958730697632,0.8462963104248047,0.8104167133569717,0.972222238779068],"occluded":false},{"timestamp":1.56,"box":[0.6395833492279053,0.8888888955116272,0.8109375238418579,0.9722222238779068],"occluded":false},{"timestamp":1.82,"box":[0.6406250596046448,0.8916666507720947,0.8145834058523178,0.972222201526165],"occluded":false},{"timestamp":2.08,"box":[0.6364583373069763,0.875,0.814583346247673,0.9722222238779068],"occluded":false},{"timestamp":2.34,"box":[0.6265625357627869,0.8287037014961243,0.8119792193174362,0.9722222238779068],"occluded":false},{"timestamp":2.6,"box":[0.6250000596046448,0.7990740537643433,0.8088542371988297,0.9722221940755844],"occluded":false},{"timestamp":2.86,"box":[0.6213542222976685,0.7675925493240356,0.8072917312383652,0.9722221791744232],"occluded":false},{"timestamp":3.12,"box":[0.6213542222976685,0.7379629611968994,0.801562562584877,0.9722222089767456],"occluded":false},{"timestamp":3.38,"box":[0.6187500357627869,0.7111110687255859,0.7911458760499954,0.9694443941116333],"occluded":false},{"timestamp":3.64,"box":[0.6078125238418579,0.6981481313705444,0.7723958641290665,0.9388888627290726],"occluded":false},{"timestamp":3.9,"box":[0.6000000238418579,0.690740704536438,0.7578125298023224,0.9222221821546555],"occluded":false},{"timestamp":4.16,"box":[0.6031250357627869,0.6824073791503906,0.7468750476837158,0.8990740478038788],"occluded":false},{"timestamp":4.42,"box":[0.6010416746139526,0.6787036657333374,0.736458346247673,0.8842592239379883],"occluded":false},{"timestamp":4.68,"box":[0.5984375476837158,0.6712962985038757,0.7286458909511566,0.868518516421318],"occluded":false},{"timestamp":4.94,"box":[0.5947917103767395,0.6675925850868225,0.7286458909511566,0.8675925880670547],"occluded":false},{"timestamp":5.2,"box":[0.5937500596046448,0.6712962985038757,0.7270833998918533,0.8694444447755814],"occluded":false},{"timestamp":5.46,"box":[0.5916666984558105,0.6759259104728699,0.7203125357627869,0.8722222000360489],"occluded":false},{"timestamp":5.72,"box":[0.5901042222976685,0.680555522441864,0.7187500596046448,0.8777777403593063],"occluded":false},{"timestamp":5.98,"box":[0.5901042222976685,0.6962962746620178,0.7171875685453415,0.8925925642251968],"occluded":false},{"timestamp":6.24,"box":[0.5895833373069763,0.710185170173645,0.7182291746139526,0.9074073880910873],"occluded":false},{"timestamp":6.5,"box":[0.5932291746139526,0.7166666388511658,0.7203125208616257,0.9111110866069794],"occluded":false},{"timestamp":6.76,"box":[0.5953125357627869,0.7055555582046509,0.7177083790302277,0.8944444507360458],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.6010416746139526,0.6944444179534912,0.7203125134110451,0.8759258985519409],"occluded":false},{"timestamp":7.28,"box":[0.6244791746139526,0.6898148059844971,0.7437500134110451,0.8675925731658936],"occluded":false},{"timestamp":7.54,"box":[0.651562511920929,0.6824073791503906,0.7697916850447655,0.8546295911073685],"occluded":false},{"timestamp":7.8,"box":[0.6786458492279053,0.6740740537643433,0.7989583536982536,0.8425925672054291],"occluded":false},{"timestamp":8.06,"box":[0.7067708969116211,0.6685184836387634,0.8265625685453415,0.8342592269182205],"occluded":false},{"timestamp":8.32,"box":[0.7333333492279053,0.675000011920929,0.8557291924953461,0.8333333432674408],"occluded":false},{"timestamp":8.58,"box":[0.7744792103767395,0.680555522441864,0.8994792103767395,0.8361110687255859],"occluded":false},{"timestamp":8.84,"box":[0.8197917342185974,0.6870370507240295,0.9526042342185974,0.8435185253620148],"occluded":false},{"timestamp":9.1,"box":[0.8682292103767395,0.6990740895271301,0.9994792193174362,0.8694444596767426],"occluded":false},{"timestamp":9.36,"box":[0.9197916984558105,0.7249999642372131,0.999479204416275,0.9046295881271362],"occluded":false},{"timestamp":9.620000000000001,"box":[0.96770840883255,0.7731481194496155,0.9994792453944683,0.9277777522802353],"occluded":false}],"scores":[0.7419355511665344,0.7419355511665344,0.7419355511665344,0.7419355511665344,0.7419355511665344,0.7419355511665344,0.7419355511665344,0.7419355511665344,0.7419355511665344,0.7419355511665344,0.7419355511665344,0.7419355511665344,0.7419355511665344,0.7419355511665344,0.7419355511665344,0.7419355511665344,0.7419355511665344,0.7419355511665344,0.7419355511665344,0.7419355511665344,0.7419355511665344,0.7419355511665344,0.7419355511665344,0.7419355511665344,0.7419355511665344,0.7419355511665344,0.7419355511665344,0.7419355511665344,0.7419355511665344,0.7419355511665344,0.7419355511665344,0.7419355511665344,0.7419355511665344,0.7419355511665344,0.7419355511665344,0.7419355511665344,0.7419355511665344,0.7419355511665344]},{"noun_phrase":"huge boulder","object_id":1,"observations":[{"timestamp":0.0,"box":[0.3203125298023224,0.6212962865829468,0.508333370089531,0.8833333253860474],"occluded":false},{"timestamp":0.26,"box":[0.32083335518836975,0.6194444298744202,0.5093750357627869,0.8814814686775208],"occluded":false},{"timestamp":0.52,"box":[0.3057291805744171,0.625,0.49791668355464935,0.8972222208976746],"occluded":false},{"timestamp":0.78,"box":[0.2744791805744171,0.654629647731781,0.48333336412906647,0.9500000178813934],"occluded":false},{"timestamp":1.04,"box":[0.25312501192092896,0.6814814805984497,0.4776041954755783,0.9712962806224823],"occluded":false},{"timestamp":1.3,"box":[0.22552084922790527,0.7444444298744202,0.4656250327825546,0.9712962806224823],"occluded":false},{"timestamp":1.56,"box":[0.22708334028720856,0.7879629731178284,0.466145858168602,0.9712963104248047],"occluded":false},{"timestamp":1.82,"box":[0.22760418057441711,0.7870370149612427,0.4671875238418579,0.9722221940755844],"occluded":false},{"timestamp":2.08,"box":[0.22708334028720856,0.7675925493240356,0.47031252086162567,0.9722221791744232],"occluded":false},{"timestamp":2.34,"box":[0.22291667759418488,0.7185184955596924,0.4687500298023224,0.9712962806224823],"occluded":false},{"timestamp":2.6,"box":[0.22291667759418488,0.6851851940155029,0.46822918951511383,0.972222238779068],"occluded":false},{"timestamp":2.86,"box":[0.22187501192092896,0.6537036895751953,0.46510419249534607,0.9712962806224823],"occluded":false},{"timestamp":3.12,"box":[0.22708334028720856,0.6277777552604675,0.46822918951511383,0.9648147821426392],"occluded":false},{"timestamp":3.38,"box":[0.2380208522081375,0.6074073910713196,0.47031253576278687,0.9324073791503906],"occluded":false},{"timestamp":3.64,"box":[0.24531251192092896,0.5981481671333313,0.46666668355464935,0.9064815044403076],"occluded":false},{"timestamp":3.9,"box":[0.2567708492279053,0.595370352268219,0.46666669845581055,0.8879629373550415],"occluded":false},{"timestamp":4.16,"box":[0.27916666865348816,0.5925925970077515,0.4755208492279053,0.8703703582286835],"occluded":false},{"timestamp":4.42,"box":[0.29322919249534607,0.5925925970077515,0.48072920739650726,0.8564814925193787],"occluded":false},{"timestamp":4.68,"box":[0.30416667461395264,0.5888888835906982,0.4828125238418579,0.8416666686534882],"occluded":false},{"timestamp":4.94,"box":[0.3020833432674408,0.5888888835906982,0.4802083522081375,0.8407407402992249],"occluded":false},{"timestamp":5.2,"box":[0.2994791865348816,0.5935184955596924,0.47812503576278687,0.8462962806224823],"occluded":false},{"timestamp":5.46,"box":[0.29739585518836975,0.5981481671333313,0.476041704416275,0.8500000238418579],"occluded":false},{"timestamp":5.72,"box":[0.29739585518836975,0.605555534362793,0.47447919845581055,0.8592592477798462],"occluded":false},{"timestamp":5.98,"box":[0.29635417461395264,0.6212962865829468,0.4739583432674408,0.875],"occluded":false},{"timestamp":6.24,"box":[0.2953125238418579,0.635185182094574,0.4734375327825546,0.8898147940635681],"occluded":false},{"timestamp":6.5,"box":[0.30520835518836975,0.6388888955116272,0.47968752682209015,0.8879629671573639],"occluded":false},{"timestamp":6.76,"box":[0.31666669249534607,0.6296296119689941,0.4854166954755783,0.8666666448116302],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.33281251788139343,0.6194444298744202,0.4947916865348816,0.849074050784111],"occluded":false},{"timestamp":7.28,"box":[0.36354169249534607,0.6129629611968994,0.5208333730697632,0.8324074000120163],"occluded":false},{"timestamp":7.54,"box":[0.40000003576278687,0.6037036776542664,0.5500000417232513,0.8111110776662827],"occluded":false},{"timestamp":7.8,"box":[0.43802085518836975,0.5962963104248047,0.5807292014360428,0.7916666716337204],"occluded":false},{"timestamp":8.06,"box":[0.47291669249534607,0.5888888835906982,0.6093750298023224,0.7759259194135666],"occluded":false},{"timestamp":8.32,"box":[0.5041667222976685,0.5907407402992249,0.6369792222976685,0.7675925940275192],"occluded":false},{"timestamp":8.58,"box":[0.5442708730697632,0.5898147821426392,0.6765625476837158,0.7601851522922516],"occluded":false},{"timestamp":8.84,"box":[0.5838541984558105,0.5888888835906982,0.7166666984558105,0.7555555403232574],"occluded":false},{"timestamp":9.1,"box":[0.6192708611488342,0.5916666388511658,0.7567708641290665,0.7564814537763596],"occluded":false},{"timestamp":9.36,"box":[0.6531250476837158,0.6027777791023254,0.7968750596046448,0.7685185223817825],"occluded":false},{"timestamp":9.620000000000001,"box":[0.6817708611488342,0.614814817905426,0.8328125327825546,0.7824074029922485],"occluded":false},{"timestamp":9.88,"box":[0.7078125476837158,0.6268518567085266,0.8671875596046448,0.7999999970197678],"occluded":false},{"timestamp":10.14,"box":[0.7494792342185974,0.6435185074806213,0.9260417371988297,0.8324074000120163],"occluded":false},{"timestamp":10.4,"box":[0.7953125238418579,0.6675925850868225,0.9932291954755783,0.8740740567445755],"occluded":false},{"timestamp":10.66,"box":[0.8447917103767395,0.6953703761100769,0.9994792193174362,0.9027777761220932],"occluded":false},{"timestamp":10.92,"box":[0.9010416865348816,0.7333333492279053,0.9994791895151138,0.9500000178813934],"occluded":false},{"timestamp":11.18,"box":[0.9718750715255737,0.805555522441864,0.9994792398065329,0.9722221791744232],"occluded":false}],"scores":[0.9179486632347107,0.9179486632347107,0.9179486632347107,0.9179486632347107,0.9179486632347107,0.9179486632347107,0.9179486632347107,0.9179486632347107,0.9179486632347107,0.9179486632347107,0.9179486632347107,0.9179486632347107,0.9179486632347107,0.9179486632347107,0.9179486632347107,0.9179486632347107,0.9179486632347107,0.9179486632347107,0.9179486632347107,0.9179486632347107,0.9179486632347107,0.9179486632347107,0.9179486632347107,0.9179486632347107,0.9179486632347107,0.9179486632347107,0.9179486632347107,0.9179486632347107,0.9179486632347107,0.9179486632347107,0.9179486632347107,0.9179486632347107,0.9179486632347107,0.9179486632347107,0.9179486632347107,0.9179486632347107,0.9179486632347107,0.9179486632347107,0.9179486632347107,0.9179486632347107,0.9179486632347107,0.9179486632347107,0.9179486632347107,0.9179486632347107]},{"noun_phrase":"huge boulder","object_id":2,"observations":[{"timestamp":0.0,"box":[0.3489583432674408,0.12037036567926407,0.7296875417232513,0.6277777925133705],"occluded":false},{"timestamp":0.26,"box":[0.3500000238418579,0.12222222238779068,0.7328125536441803,0.6268518343567848],"occluded":false},{"timestamp":0.52,"box":[0.33385419845581055,0.11574073880910873,0.7229167222976685,0.6342592462897301],"occluded":false},{"timestamp":0.78,"box":[0.3083333373069763,0.11296296119689941,0.7229166924953461,0.664814829826355],"occluded":false},{"timestamp":1.04,"box":[0.2848958373069763,0.09537036716938019,0.7322916984558105,0.6916666775941849],"occluded":false},{"timestamp":1.3,"box":[0.26093751192092896,0.12037036567926407,0.736979216337204,0.756481446325779],"occluded":false},{"timestamp":1.56,"box":[0.25416669249534607,0.14814814925193787,0.7421875596046448,0.8009259402751923],"occluded":false},{"timestamp":1.82,"box":[0.2562500238418579,0.14999999105930328,0.7427083849906921,0.7999999672174454],"occluded":false},{"timestamp":2.08,"box":[0.2593750059604645,0.13148148357868195,0.7468750178813934,0.781481459736824],"occluded":false},{"timestamp":2.34,"box":[0.2588541805744171,0.0833333283662796,0.7463541924953461,0.7314814478158951],"occluded":false},{"timestamp":2.6,"box":[0.2604166865348816,0.05000000074505806,0.7505208849906921,0.6972222216427326],"occluded":false},{"timestamp":2.86,"box":[0.2588541805744171,0.031481482088565826,0.749479204416275,0.6685185208916664],"occluded":false},{"timestamp":3.12,"box":[0.2614583373069763,0.031481482088565826,0.7463541924953461,0.6425925865769386],"occluded":false},{"timestamp":3.38,"box":[0.27031251788139343,0.031481482088565826,0.7390625476837158,0.6203703656792641],"occluded":false},{"timestamp":3.64,"box":[0.27656251192092896,0.03055555559694767,0.7250000238418579,0.6092592570930719],"occluded":false},{"timestamp":3.9,"box":[0.2864583432674408,0.0416666641831398,0.7098958790302277,0.6064814701676369],"occluded":false},{"timestamp":4.16,"box":[0.30677086114883423,0.069444440305233,0.705729216337204,0.5999999865889549],"occluded":false},{"timestamp":4.42,"box":[0.3192708492279053,0.0962962955236435,0.6989583671092987,0.6018518656492233],"occluded":false},{"timestamp":4.68,"box":[0.32864585518836975,0.11203703284263611,0.6921875476837158,0.5962962806224823],"occluded":false},{"timestamp":4.94,"box":[0.3265625238418579,0.11481481045484543,0.6875000298023224,0.5962962731719017],"occluded":false},{"timestamp":5.2,"box":[0.32343751192092896,0.11759258806705475,0.6843750178813934,0.5999999791383743],"occluded":false},{"timestamp":5.46,"box":[0.3218750059604645,0.1240740716457367,0.6822916865348816,0.6027777791023254],"occluded":false},{"timestamp":5.72,"box":[0.31979167461395264,0.13240741193294525,0.679166704416275,0.6092592626810074],"occluded":false},{"timestamp":5.98,"box":[0.31979167461395264,0.14907407760620117,0.6786458492279053,0.6268518567085266],"occluded":false},{"timestamp":6.24,"box":[0.31979167461395264,0.16388888657093048,0.679166704416275,0.6407407373189926],"occluded":false},{"timestamp":6.5,"box":[0.32864585518836975,0.17777776718139648,0.6817708611488342,0.6472221910953522],"occluded":false},{"timestamp":6.76,"box":[0.3411458432674408,0.18425925076007843,0.6822916865348816,0.6370370239019394],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.3552083373069763,0.18888889253139496,0.6854166984558105,0.6259259134531021],"occluded":false},{"timestamp":7.28,"box":[0.38697919249534607,0.19629628956317902,0.7072917222976685,0.6203703433275223],"occluded":false},{"timestamp":7.54,"box":[0.4229166805744171,0.20277777314186096,0.7333333492279053,0.6111111044883728],"occluded":false},{"timestamp":7.8,"box":[0.460416704416275,0.21203702688217163,0.760416716337204,0.6064814627170563],"occluded":false},{"timestamp":8.06,"box":[0.4947916865348816,0.2203703671693802,0.7890625298023224,0.599074050784111],"occluded":false},{"timestamp":8.32,"box":[0.526562511920929,0.2342592477798462,0.8161458671092987,0.6009259223937988],"occluded":false},{"timestamp":8.58,"box":[0.5677083730697632,0.24074073135852814,0.861979216337204,0.6018518358469009],"occluded":false},{"timestamp":8.84,"box":[0.6072916984558105,0.23888888955116272,0.9151042103767395,0.6027777791023254],"occluded":false},{"timestamp":9.1,"box":[0.6453125476837158,0.2370370328426361,0.9739584028720856,0.6092592477798462],"occluded":false},{"timestamp":9.36,"box":[0.6791666746139526,0.2361111044883728,0.999479204416275,0.6222222149372101],"occluded":false},{"timestamp":9.620000000000001,"box":[0.7098958492279053,0.2361111044883728,0.999479204416275,0.6370370090007782],"occluded":false},{"timestamp":9.88,"box":[0.7364583611488342,0.2361111044883728,0.999479204416275,0.8870370388031006],"occluded":false},{"timestamp":10.14,"box":[0.7807292342185974,0.23148147761821747,0.9994792491197586,0.6768518537282944],"occluded":false},{"timestamp":10.4,"box":[0.8281250596046448,0.2481481432914734,0.9994792342185974,0.7064814567565918],"occluded":false},{"timestamp":10.66,"box":[0.8822917342185974,0.27314814925193787,0.999479241669178,0.710185170173645],"occluded":false},{"timestamp":10.92,"box":[0.9375000596046448,0.32499998807907104,0.9994792304933071,0.8592592477798462],"occluded":false}],"scores":[0.9532163739204407,0.9532163739204407,0.9532163739204407,0.9532163739204407,0.9532163739204407,0.9532163739204407,0.9532163739204407,0.9532163739204407,0.9532163739204407,0.9532163739204407,0.9532163739204407,0.9532163739204407,0.9532163739204407,0.9532163739204407,0.9532163739204407,0.9532163739204407,0.9532163739204407,0.9532163739204407,0.9532163739204407,0.9532163739204407,0.9532163739204407,0.9532163739204407,0.9532163739204407,0.9532163739204407,0.9532163739204407,0.9532163739204407,0.9532163739204407,0.9532163739204407,0.9532163739204407,0.9532163739204407,0.9532163739204407,0.9532163739204407,0.9532163739204407,0.9532163739204407,0.9532163739204407,0.9532163739204407,0.9532163739204407,0.9532163739204407,0.9532163739204407,0.9532163739204407,0.9532163739204407,0.9532163739204407,0.9532163739204407]},{"noun_phrase":"red door","object_id":0,"observations":[{"timestamp":0.0,"box":[0.5531250238418579,0.7064814567565918,0.610416691750288,0.8157407119870186],"occluded":false},{"timestamp":0.26,"box":[0.5536458492279053,0.7046296000480652,0.6125000193715096,0.8157407119870186],"occluded":false},{"timestamp":0.52,"box":[0.01875000074505806,0.7120370268821716,0.603125024586916,0.9842592477798462],"occluded":false},{"timestamp":0.78,"box":[0.004687500186264515,0.7481481432914734,0.5953125478699803,0.9990740716457367],"occluded":false},{"timestamp":1.04,"box":[0.01875000074505806,0.7787036895751953,0.5979167111217976,0.9990740567445755],"occluded":false},{"timestamp":1.3,"box":[0.07447917014360428,0.8490740656852722,0.5942708924412727,0.9990740567445755],"occluded":false},{"timestamp":1.56,"box":[0.07447917014360428,0.8907407522201538,0.5989583805203438,0.9990740865468979],"occluded":false},{"timestamp":1.82,"box":[0.06041666865348816,0.8907407522201538,0.6000000536441803,0.9990740865468979],"occluded":false},{"timestamp":2.08,"box":[0.5260416865348816,0.8740740418434143,0.6015625223517418,0.9722221866250038],"occluded":false},{"timestamp":2.34,"box":[0.5250000357627869,0.824999988079071,0.6000000387430191,0.970370352268219],"occluded":false},{"timestamp":2.6,"box":[0.5244792103767395,0.7935184836387634,0.5989583805203438,0.9361110776662827],"occluded":false},{"timestamp":2.86,"box":[0.5218750238418579,0.7638888955116272,0.5958333611488342,0.9046296328306198],"occluded":false},{"timestamp":3.12,"box":[0.5234375,0.7342592477798462,0.5963541716337204,0.8731481283903122],"occluded":false},{"timestamp":3.38,"box":[0.5234375,0.7092592716217041,0.5942708402872086,0.8425925970077515],"occluded":false},{"timestamp":3.64,"box":[0.518750011920929,0.6953703761100769,0.5859375149011612,0.8222222328186035],"occluded":false},{"timestamp":3.9,"box":[0.5151041746139526,0.6879629492759705,0.5786458477377892,0.8083333149552345],"occluded":false},{"timestamp":4.16,"box":[0.5218750238418579,0.6814814805984497,0.5822916924953461,0.7944444417953491],"occluded":false},{"timestamp":4.42,"box":[0.5234375,0.6759259104728699,0.5812500044703484,0.784259244799614],"occluded":false},{"timestamp":4.68,"box":[0.02187500149011612,0.6685184836387634,0.5807292237877846,0.9629629254341125],"occluded":false},{"timestamp":4.94,"box":[0.01875000074505806,0.6675925850868225,0.576562512665987,0.9351851642131805],"occluded":false},{"timestamp":5.2,"box":[0.5197917222976685,0.67037034034729,0.5755208916962147,0.7731481194496155],"occluded":false},{"timestamp":5.46,"box":[0.03281250223517418,0.675000011920929,0.5723958872258663,0.9351851940155029],"occluded":false},{"timestamp":5.72,"box":[0.5166667103767395,0.680555522441864,0.5713542141020298,0.7833333015441895],"occluded":false},{"timestamp":5.98,"box":[0.515625,0.6972222328186035,0.5713541693985462,0.800000011920929],"occluded":false},{"timestamp":6.24,"box":[0.004687500186264515,0.7120370268821716,0.5703125121071935,0.9842592477798462],"occluded":false},{"timestamp":6.5,"box":[0.03281250223517418,0.7138888835906982,0.5755208395421505,0.9842592477798462],"occluded":false},{"timestamp":6.76,"box":[0.0598958358168602,0.7037037014961243,0.5770833715796471,0.970370352268219],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.08750000596046448,0.6925925612449646,0.5828125178813934,0.9425925612449646],"occluded":false},{"timestamp":7.28,"box":[0.12968750298023224,0.6851851940155029,0.6072916835546494,0.9148148149251938],"occluded":false},{"timestamp":7.54,"box":[0.1848958432674408,0.6759259104728699,0.6338542103767395,0.8731481283903122],"occluded":false},{"timestamp":7.8,"box":[0.24114584922790527,0.6666666269302368,0.6619791984558105,0.8592592179775238],"occluded":false},{"timestamp":8.06,"box":[0.2828125059604645,0.6601851582527161,0.6895833611488342,0.859259232878685],"occluded":false},{"timestamp":8.32,"box":[0.33802086114883423,0.6601851582527161,0.7166667282581329,0.7898147851228714],"occluded":false},{"timestamp":8.58,"box":[0.7098958492279053,0.6611111164093018,0.7557291835546494,0.7453703731298447],"occluded":false},{"timestamp":8.84,"box":[0.7515625357627869,0.6629629731178284,0.800000037997961,0.7500000074505806],"occluded":false},{"timestamp":9.1,"box":[0.7937500476837158,0.67037034034729,0.8453125506639481,0.7620370090007782],"occluded":false},{"timestamp":9.36,"box":[0.8364583849906921,0.6861110925674438,0.8937500528991222,0.7833333164453506],"occluded":false},{"timestamp":9.620000000000001,"box":[0.8760417103767395,0.7046296000480652,0.9380208812654018,0.8083333000540733],"occluded":false},{"timestamp":9.88,"box":[0.9135417342185974,0.7240740656852722,0.9812500700354576,0.8351851776242256],"occluded":false},{"timestamp":10.14,"box":[0.9781250357627869,0.7527777552604675,0.999479204416275,0.7842592373490334],"occluded":false}],"scores":[0.8324607610702515,0.8324607610702515,0.8324607610702515,0.8324607610702515,0.8324607610702515,0.8324607610702515,0.8324607610702515,0.8324607610702515,0.8324607610702515,0.8324607610702515,0.8324607610702515,0.8324607610702515,0.8324607610702515,0.8324607610702515,0.8324607610702515,0.8324607610702515,0.8324607610702515,0.8324607610702515,0.8324607610702515,0.8324607610702515,0.8324607610702515,0.8324607610702515,0.8324607610702515,0.8324607610702515,0.8324607610702515,0.8324607610702515,0.8324607610702515,0.8324607610702515,0.8324607610702515,0.8324607610702515,0.8324607610702515,0.8324607610702515,0.8324607610702515,0.8324607610702515,0.8324607610702515,0.8324607610702515,0.8324607610702515,0.8324607610702515,0.8324607610702515,0.8324607610702515]},{"noun_phrase":"red door","object_id":2,"observations":[{"timestamp":6.76,"box":[0.0364583358168602,0.9416666626930237,0.07604167237877846,0.9722222182899714],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.06875000149011612,0.9138888716697693,0.10468750447034836,0.9722222052514553],"occluded":false},{"timestamp":7.28,"box":[0.11875000596046448,0.8879629373550415,0.15312500670552254,0.9712962657213211],"occluded":false},{"timestamp":7.54,"box":[0.1744791716337204,0.8518518209457397,0.20468750596046448,0.9537036716938019],"occluded":false},{"timestamp":7.8,"box":[0.23072917759418488,0.8203703761100769,0.25885417871177197,0.9138888940215111],"occluded":false},{"timestamp":8.06,"box":[0.28125,0.79537034034729,0.30572916753590107,0.879629597067833],"occluded":false},{"timestamp":8.32,"box":[0.32343751192092896,0.7814814448356628,0.346354179084301,0.8583332970738411],"occluded":false},{"timestamp":8.58,"box":[0.3739583492279053,0.7638888955116272,0.3958333507180214,0.8361111134290695],"occluded":false},{"timestamp":8.84,"box":[0.41822919249534607,0.7518518567085266,0.4375000260770321,0.8203703761100769],"occluded":false},{"timestamp":9.1,"box":[0.4557291865348816,0.7462962865829468,0.4760416876524687,0.8120370283722878],"occluded":false},{"timestamp":9.36,"box":[0.48593753576278687,0.7527777552604675,0.505729204043746,0.8185184970498085],"occluded":false},{"timestamp":9.620000000000001,"box":[0.510937511920929,0.760185182094574,0.531770845875144,0.825925923883915],"occluded":false},{"timestamp":9.88,"box":[0.5328125357627869,0.7694444060325623,0.552604204043746,0.8351851478219032],"occluded":false},{"timestamp":10.14,"box":[0.5671875476837158,0.7814814448356628,0.5875000488013029,0.8462962582707405],"occluded":false},{"timestamp":10.4,"box":[0.6015625596046448,0.7990740537643433,0.99114590883255,0.9990740567445755],"occluded":false},{"timestamp":10.66,"box":[0.6380208730697632,0.8185185194015503,0.7416667118668556,0.9990740716457367],"occluded":false},{"timestamp":10.92,"box":[0.6755208969116211,0.8509259223937988,0.9354167580604553,0.9990740716457367],"occluded":false},{"timestamp":11.18,"box":[0.7177083492279053,0.8907407522201538,0.7427083514630795,0.9685185253620148],"occluded":false},{"timestamp":11.44,"box":[0.7697917222976685,0.9296296238899231,0.7963542230427265,0.9722222164273262],"occluded":false},{"timestamp":12.74,"box":[0.9171875715255737,0.9444444179534912,0.953125074505806,0.9731481205672026],"occluded":false}],"scores":[0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115]}],"tracklets":[{"noun_phrase":"huge boulder","object_id":0,"presentation_id":1,"observations":[{"timestamp":0.0,"box":[0.6322916746139526,0.7046296000480652,0.7744791805744171,0.9231481105089188],"occluded":false},{"timestamp":0.26,"box":[0.6328125596046448,0.7064814567565918,0.7739583998918533,0.9212962687015533],"occluded":false},{"timestamp":0.52,"box":[0.6244791746139526,0.7111110687255859,0.768750011920929,0.9305555075407028],"occluded":false},{"timestamp":0.78,"box":[0.6177083849906921,0.75,0.7734375596046448,0.9703703671693802],"occluded":false},{"timestamp":1.04,"box":[0.6223958730697632,0.7768518328666687,0.7875000536441803,0.9722221940755844],"occluded":false},{"timestamp":1.3,"box":[0.6223958730697632,0.8462963104248047,0.8104167133569717,0.972222238779068],"occluded":false},{"timestamp":1.56,"box":[0.6395833492279053,0.8888888955116272,0.8109375238418579,0.9722222238779068],"occluded":false},{"timestamp":1.82,"box":[0.6406250596046448,0.8916666507720947,0.8145834058523178,0.972222201526165],"occluded":false},{"timestamp":2.08,"box":[0.6364583373069763,0.875,0.814583346247673,0.9722222238779068],"occluded":false},{"timestamp":2.34,"box":[0.6265625357627869,0.8287037014961243,0.8119792193174362,0.9722222238779068],"occluded":false},{"timestamp":2.6,"box":[0.6250000596046448,0.7990740537643433,0.8088542371988297,0.9722221940755844],"occluded":false},{"timestamp":2.86,"box":[0.6213542222976685,0.7675925493240356,0.8072917312383652,0.9722221791744232],"occluded":false},{"timestamp":3.12,"box":[0.6213542222976685,0.7379629611968994,0.801562562584877,0.9722222089767456],"occluded":false},{"timestamp":3.38,"box":[0.6187500357627869,0.7111110687255859,0.7911458760499954,0.9694443941116333],"occluded":false},{"timestamp":3.64,"box":[0.6078125238418579,0.6981481313705444,0.7723958641290665,0.9388888627290726],"occluded":false},{"timestamp":3.9,"box":[0.6000000238418579,0.690740704536438,0.7578125298023224,0.9222221821546555],"occluded":false},{"timestamp":4.16,"box":[0.6031250357627869,0.6824073791503906,0.7468750476837158,0.8990740478038788],"occluded":false},{"timestamp":4.42,"box":[0.6010416746139526,0.6787036657333374,0.736458346247673,0.8842592239379883],"occluded":false},{"timestamp":4.68,"box":[0.5984375476837158,0.6712962985038757,0.7286458909511566,0.868518516421318],"occluded":false},{"timestamp":4.94,"box":[0.5947917103767395,0.6675925850868225,0.7286458909511566,0.8675925880670547],"occluded":false},{"timestamp":5.2,"box":[0.5937500596046448,0.6712962985038757,0.7270833998918533,0.8694444447755814],"occluded":false},{"timestamp":5.46,"box":[0.5916666984558105,0.6759259104728699,0.7203125357627869,0.8722222000360489],"occluded":false},{"timestamp":5.72,"box":[0.5901042222976685,0.680555522441864,0.7187500596046448,0.8777777403593063],"occluded":false},{"timestamp":5.98,"box":[0.5901042222976685,0.6962962746620178,0.7171875685453415,0.8925925642251968],"occluded":false},{"timestamp":6.24,"box":[0.5895833373069763,0.710185170173645,0.7182291746139526,0.9074073880910873],"occluded":false},{"timestamp":6.5,"box":[0.5932291746139526,0.7166666388511658,0.7203125208616257,0.9111110866069794],"occluded":false},{"timestamp":6.76,"box":[0.5953125357627869,0.7055555582046509,0.7177083790302277,0.8944444507360458],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.6010416746139526,0.6944444179534912,0.7203125134110451,0.8759258985519409],"occluded":false},{"timestamp":7.28,"box":[0.6244791746139526,0.6898148059844971,0.7437500134110451,0.8675925731658936],"occluded":false},{"timestamp":7.54,"box":[0.651562511920929,0.6824073791503906,0.7697916850447655,0.8546295911073685],"occluded":false},{"timestamp":7.8,"box":[0.6786458492279053,0.6740740537643433,0.7989583536982536,0.8425925672054291],"occluded":false},{"timestamp":8.06,"box":[0.7067708969116211,0.6685184836387634,0.8265625685453415,0.8342592269182205],"occluded":false},{"timestamp":8.32,"box":[0.7333333492279053,0.675000011920929,0.8557291924953461,0.8333333432674408],"occluded":false},{"timestamp":8.58,"box":[0.7744792103767395,0.680555522441864,0.8994792103767395,0.8361110687255859],"occluded":false},{"timestamp":8.84,"box":[0.8197917342185974,0.6870370507240295,0.9526042342185974,0.8435185253620148],"occluded":false},{"timestamp":9.1,"box":[0.8682292103767395,0.6990740895271301,0.9994792193174362,0.8694444596767426],"occluded":false},{"timestamp":9.36,"box":[0.9197916984558105,0.7249999642372131,0.999479204416275,0.9046295881271362],"occluded":false},{"timestamp":9.620000000000001,"box":[0.96770840883255,0.7731481194496155,0.9994792453944683,0.9277777522802353],"occluded":false}],"mean_score":0.7419355511665344},{"noun_phrase":"red door","object_id":2,"presentation_id":2,"observations":[{"timestamp":6.76,"box":[0.0364583358168602,0.9416666626930237,0.07604167237877846,0.9722222182899714],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.06875000149011612,0.9138888716697693,0.10468750447034836,0.9722222052514553],"occluded":false},{"timestamp":7.28,"box":[0.11875000596046448,0.8879629373550415,0.15312500670552254,0.9712962657213211],"occluded":false},{"timestamp":7.54,"box":[0.1744791716337204,0.8518518209457397,0.20468750596046448,0.9537036716938019],"occluded":false},{"timestamp":7.8,"box":[0.23072917759418488,0.8203703761100769,0.25885417871177197,0.9138888940215111],"occluded":false},{"timestamp":8.06,"box":[0.28125,0.79537034034729,0.30572916753590107,0.879629597067833],"occluded":false},{"timestamp":8.32,"box":[0.32343751192092896,0.7814814448356628,0.346354179084301,0.8583332970738411],"occluded":false},{"timestamp":8.58,"box":[0.3739583492279053,0.7638888955116272,0.3958333507180214,0.8361111134290695],"occluded":false},{"timestamp":8.84,"box":[0.41822919249534607,0.7518518567085266,0.4375000260770321,0.8203703761100769],"occluded":false},{"timestamp":9.1,"box":[0.4557291865348816,0.7462962865829468,0.4760416876524687,0.8120370283722878],"occluded":false},{"timestamp":9.36,"box":[0.48593753576278687,0.7527777552604675,0.505729204043746,0.8185184970498085],"occluded":false},{"timestamp":9.620000000000001,"box":[0.510937511920929,0.760185182094574,0.531770845875144,0.825925923883915],"occluded":false},{"timestamp":9.88,"box":[0.5328125357627869,0.7694444060325623,0.552604204043746,0.8351851478219032],"occluded":false},{"timestamp":10.14,"box":[0.5671875476837158,0.7814814448356628,0.5875000488013029,0.8462962582707405],"occluded":false},{"timestamp":10.4,"box":[0.6015625596046448,0.7990740537643433,0.99114590883255,0.9990740567445755],"occluded":false},{"timestamp":10.66,"box":[0.6380208730697632,0.8185185194015503,0.7416667118668556,0.9990740716457367],"occluded":false},{"timestamp":10.92,"box":[0.6755208969116211,0.8509259223937988,0.9354167580604553,0.9990740716457367],"occluded":false},{"timestamp":11.18,"box":[0.7177083492279053,0.8907407522201538,0.7427083514630795,0.9685185253620148],"occluded":false},{"timestamp":11.44,"box":[0.7697917222976685,0.9296296238899231,0.7963542230427265,0.9722222164273262],"occluded":false},{"timestamp":12.74,"box":[0.9171875715255737,0.9444444179534912,0.953125074505806,0.9731481205672026],"occluded":false}],"mean_score":0.9139785170555115},{"noun_phrase":"red door","object_id":0,"presentation_id":3,"observations":[{"timestamp":0.0,"box":[0.5531250238418579,0.7064814567565918,0.610416691750288,0.8157407119870186],"occluded":false},{"timestamp":0.26,"box":[0.5536458492279053,0.7046296000480652,0.6125000193715096,0.8157407119870186],"occluded":false},{"timestamp":0.52,"box":[0.01875000074505806,0.7120370268821716,0.603125024586916,0.9842592477798462],"occluded":false},{"timestamp":0.78,"box":[0.004687500186264515,0.7481481432914734,0.5953125478699803,0.9990740716457367],"occluded":false},{"timestamp":1.04,"box":[0.01875000074505806,0.7787036895751953,0.5979167111217976,0.9990740567445755],"occluded":false},{"timestamp":1.3,"box":[0.07447917014360428,0.8490740656852722,0.5942708924412727,0.9990740567445755],"occluded":false},{"timestamp":1.56,"box":[0.07447917014360428,0.8907407522201538,0.5989583805203438,0.9990740865468979],"occluded":false},{"timestamp":1.82,"box":[0.06041666865348816,0.8907407522201538,0.6000000536441803,0.9990740865468979],"occluded":false},{"timestamp":2.08,"box":[0.5260416865348816,0.8740740418434143,0.6015625223517418,0.9722221866250038],"occluded":false},{"timestamp":2.34,"box":[0.5250000357627869,0.824999988079071,0.6000000387430191,0.970370352268219],"occluded":false},{"timestamp":2.6,"box":[0.5244792103767395,0.7935184836387634,0.5989583805203438,0.9361110776662827],"occluded":false},{"timestamp":2.86,"box":[0.5218750238418579,0.7638888955116272,0.5958333611488342,0.9046296328306198],"occluded":false},{"timestamp":3.12,"box":[0.5234375,0.7342592477798462,0.5963541716337204,0.8731481283903122],"occluded":false},{"timestamp":3.38,"box":[0.5234375,0.7092592716217041,0.5942708402872086,0.8425925970077515],"occluded":false},{"timestamp":3.64,"box":[0.518750011920929,0.6953703761100769,0.5859375149011612,0.8222222328186035],"occluded":false},{"timestamp":3.9,"box":[0.5151041746139526,0.6879629492759705,0.5786458477377892,0.8083333149552345],"occluded":false},{"timestamp":4.16,"box":[0.5218750238418579,0.6814814805984497,0.5822916924953461,0.7944444417953491],"occluded":false},{"timestamp":4.42,"box":[0.5234375,0.6759259104728699,0.5812500044703484,0.784259244799614],"occluded":false},{"timestamp":4.68,"box":[0.02187500149011612,0.6685184836387634,0.5807292237877846,0.9629629254341125],"occluded":false},{"timestamp":4.94,"box":[0.01875000074505806,0.6675925850868225,0.576562512665987,0.9351851642131805],"occluded":false},{"timestamp":5.2,"box":[0.5197917222976685,0.67037034034729,0.5755208916962147,0.7731481194496155],"occluded":false},{"timestamp":5.46,"box":[0.03281250223517418,0.675000011920929,0.5723958872258663,0.9351851940155029],"occluded":false},{"timestamp":5.72,"box":[0.5166667103767395,0.680555522441864,0.5713542141020298,0.7833333015441895],"occluded":false},{"timestamp":5.98,"box":[0.515625,0.6972222328186035,0.5713541693985462,0.800000011920929],"occluded":false},{"timestamp":6.24,"box":[0.004687500186264515,0.7120370268821716,0.5703125121071935,0.9842592477798462],"occluded":false},{"timestamp":6.5,"box":[0.03281250223517418,0.7138888835906982,0.5755208395421505,0.9842592477798462],"occluded":false},{"timestamp":6.76,"box":[0.0598958358168602,0.7037037014961243,0.5770833715796471,0.970370352268219],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.08750000596046448,0.6925925612449646,0.5828125178813934,0.9425925612449646],"occluded":false},{"timestamp":7.28,"box":[0.12968750298023224,0.6851851940155029,0.6072916835546494,0.9148148149251938],"occluded":false},{"timestamp":7.54,"box":[0.1848958432674408,0.6759259104728699,0.6338542103767395,0.8731481283903122],"occluded":false},{"timestamp":7.8,"box":[0.24114584922790527,0.6666666269302368,0.6619791984558105,0.8592592179775238],"occluded":false},{"timestamp":8.06,"box":[0.2828125059604645,0.6601851582527161,0.6895833611488342,0.859259232878685],"occluded":false},{"timestamp":8.32,"box":[0.33802086114883423,0.6601851582527161,0.7166667282581329,0.7898147851228714],"occluded":false},{"timestamp":8.58,"box":[0.7098958492279053,0.6611111164093018,0.7557291835546494,0.7453703731298447],"occluded":false},{"timestamp":8.84,"box":[0.7515625357627869,0.6629629731178284,0.800000037997961,0.7500000074505806],"occluded":false},{"timestamp":9.1,"box":[0.7937500476837158,0.67037034034729,0.8453125506639481,0.7620370090007782],"occluded":false},{"timestamp":9.36,"box":[0.8364583849906921,0.6861110925674438,0.8937500528991222,0.7833333164453506],"occluded":false},{"timestamp":9.620000000000001,"box":[0.8760417103767395,0.7046296000480652,0.9380208812654018,0.8083333000540733],"occluded":false},{"timestamp":9.88,"box":[0.9135417342185974,0.7240740656852722,0.9812500700354576,0.8351851776242256],"occluded":false},{"timestamp":10.14,"box":[0.9781250357627869,0.7527777552604675,0.999479204416275,0.7842592373490334],"occluded":false}],"mean_score":0.8324607610702515},{"noun_phrase":"huge boulder","object_id":2,"presentation_id":4,"observations":[{"timestamp":0.0,"box":[0.3489583432674408,0.12037036567926407,0.7296875417232513,0.6277777925133705],"occluded":false},{"timestamp":0.26,"box":[0.3500000238418579,0.12222222238779068,0.7328125536441803,0.6268518343567848],"occluded":false},{"timestamp":0.52,"box":[0.33385419845581055,0.11574073880910873,0.7229167222976685,0.6342592462897301],"occluded":false},{"timestamp":0.78,"box":[0.3083333373069763,0.11296296119689941,0.7229166924953461,0.664814829826355],"occluded":false},{"timestamp":1.04,"box":[0.2848958373069763,0.09537036716938019,0.7322916984558105,0.6916666775941849],"occluded":false},{"timestamp":1.3,"box":[0.26093751192092896,0.12037036567926407,0.736979216337204,0.756481446325779],"occluded":false},{"timestamp":1.56,"box":[0.25416669249534607,0.14814814925193787,0.7421875596046448,0.8009259402751923],"occluded":false},{"timestamp":1.82,"box":[0.2562500238418579,0.14999999105930328,0.7427083849906921,0.7999999672174454],"occluded":false},{"timestamp":2.08,"box":[0.2593750059604645,0.13148148357868195,0.7468750178813934,0.781481459736824],"occluded":false},{"timestamp":2.34,"box":[0.2588541805744171,0.0833333283662796,0.7463541924953461,0.7314814478158951],"occluded":false},{"timestamp":2.6,"box":[0.2604166865348816,0.05000000074505806,0.7505208849906921,0.6972222216427326],"occluded":false},{"timestamp":2.86,"box":[0.2588541805744171,0.031481482088565826,0.749479204416275,0.6685185208916664],"occluded":false},{"timestamp":3.12,"box":[0.2614583373069763,0.031481482088565826,0.7463541924953461,0.6425925865769386],"occluded":false},{"timestamp":3.38,"box":[0.27031251788139343,0.031481482088565826,0.7390625476837158,0.6203703656792641],"occluded":false},{"timestamp":3.64,"box":[0.27656251192092896,0.03055555559694767,0.7250000238418579,0.6092592570930719],"occluded":false},{"timestamp":3.9,"box":[0.2864583432674408,0.0416666641831398,0.7098958790302277,0.6064814701676369],"occluded":false},{"timestamp":4.16,"box":[0.30677086114883423,0.069444440305233,0.705729216337204,0.5999999865889549],"occluded":false},{"timestamp":4.42,"box":[0.3192708492279053,0.0962962955236435,0.6989583671092987,0.6018518656492233],"occluded":false},{"timestamp":4.68,"box":[0.32864585518836975,0.11203703284263611,0.6921875476837158,0.5962962806224823],"occluded":false},{"timestamp":4.94,"box":[0.3265625238418579,0.11481481045484543,0.6875000298023224,0.5962962731719017],"occluded":false},{"timestamp":5.2,"box":[0.32343751192092896,0.11759258806705475,0.6843750178813934,0.5999999791383743],"occluded":false},{"timestamp":5.46,"box":[0.3218750059604645,0.1240740716457367,0.6822916865348816,0.6027777791023254],"occluded":false},{"timestamp":5.72,"box":[0.31979167461395264,0.13240741193294525,0.679166704416275,0.6092592626810074],"occluded":false},{"timestamp":5.98,"box":[0.31979167461395264,0.14907407760620117,0.6786458492279053,0.6268518567085266],"occluded":false},{"timestamp":6.24,"box":[0.31979167461395264,0.16388888657093048,0.679166704416275,0.6407407373189926],"occluded":false},{"timestamp":6.5,"box":[0.32864585518836975,0.17777776718139648,0.6817708611488342,0.6472221910953522],"occluded":false},{"timestamp":6.76,"box":[0.3411458432674408,0.18425925076007843,0.6822916865348816,0.6370370239019394],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.3552083373069763,0.18888889253139496,0.6854166984558105,0.6259259134531021],"occluded":false},{"timestamp":7.28,"box":[0.38697919249534607,0.19629628956317902,0.7072917222976685,0.6203703433275223],"occluded":false},{"timestamp":7.54,"box":[0.4229166805744171,0.20277777314186096,0.7333333492279053,0.6111111044883728],"occluded":false},{"timestamp":7.8,"box":[0.460416704416275,0.21203702688217163,0.760416716337204,0.6064814627170563],"occluded":false},{"timestamp":8.06,"box":[0.4947916865348816,0.2203703671693802,0.7890625298023224,0.599074050784111],"occluded":false},{"timestamp":8.32,"box":[0.526562511920929,0.2342592477798462,0.8161458671092987,0.6009259223937988],"occluded":false},{"timestamp":8.58,"box":[0.5677083730697632,0.24074073135852814,0.861979216337204,0.6018518358469009],"occluded":false},{"timestamp":8.84,"box":[0.6072916984558105,0.23888888955116272,0.9151042103767395,0.6027777791023254],"occluded":false},{"timestamp":9.1,"box":[0.6453125476837158,0.2370370328426361,0.9739584028720856,0.6092592477798462],"occluded":false},{"timestamp":9.36,"box":[0.6791666746139526,0.2361111044883728,0.999479204416275,0.6222222149372101],"occluded":false},{"timestamp":9.620000000000001,"box":[0.7098958492279053,0.2361111044883728,0.999479204416275,0.6370370090007782],"occluded":false},{"timestamp":9.88,"box":[0.7364583611488342,0.2361111044883728,0.999479204416275,0.8870370388031006],"occluded":false},{"timestamp":10.14,"box":[0.7807292342185974,0.23148147761821747,0.9994792491197586,0.6768518537282944],"occluded":false},{"timestamp":10.4,"box":[0.8281250596046448,0.2481481432914734,0.9994792342185974,0.7064814567565918],"occluded":false},{"timestamp":10.66,"box":[0.8822917342185974,0.27314814925193787,0.999479241669178,0.710185170173645],"occluded":false},{"timestamp":10.92,"box":[0.9375000596046448,0.32499998807907104,0.9994792304933071,0.8592592477798462],"occluded":false}],"mean_score":0.9532163739204407},{"noun_phrase":"huge boulder","object_id":1,"presentation_id":5,"observations":[{"timestamp":0.0,"box":[0.3203125298023224,0.6212962865829468,0.508333370089531,0.8833333253860474],"occluded":false},{"timestamp":0.26,"box":[0.32083335518836975,0.6194444298744202,0.5093750357627869,0.8814814686775208],"occluded":false},{"timestamp":0.52,"box":[0.3057291805744171,0.625,0.49791668355464935,0.8972222208976746],"occluded":false},{"timestamp":0.78,"box":[0.2744791805744171,0.654629647731781,0.48333336412906647,0.9500000178813934],"occluded":false},{"timestamp":1.04,"box":[0.25312501192092896,0.6814814805984497,0.4776041954755783,0.9712962806224823],"occluded":false},{"timestamp":1.3,"box":[0.22552084922790527,0.7444444298744202,0.4656250327825546,0.9712962806224823],"occluded":false},{"timestamp":1.56,"box":[0.22708334028720856,0.7879629731178284,0.466145858168602,0.9712963104248047],"occluded":false},{"timestamp":1.82,"box":[0.22760418057441711,0.7870370149612427,0.4671875238418579,0.9722221940755844],"occluded":false},{"timestamp":2.08,"box":[0.22708334028720856,0.7675925493240356,0.47031252086162567,0.9722221791744232],"occluded":false},{"timestamp":2.34,"box":[0.22291667759418488,0.7185184955596924,0.4687500298023224,0.9712962806224823],"occluded":false},{"timestamp":2.6,"box":[0.22291667759418488,0.6851851940155029,0.46822918951511383,0.972222238779068],"occluded":false},{"timestamp":2.86,"box":[0.22187501192092896,0.6537036895751953,0.46510419249534607,0.9712962806224823],"occluded":false},{"timestamp":3.12,"box":[0.22708334028720856,0.6277777552604675,0.46822918951511383,0.9648147821426392],"occluded":false},{"timestamp":3.38,"box":[0.2380208522081375,0.6074073910713196,0.47031253576278687,0.9324073791503906],"occluded":false},{"timestamp":3.64,"box":[0.24531251192092896,0.5981481671333313,0.46666668355464935,0.9064815044403076],"occluded":false},{"timestamp":3.9,"box":[0.2567708492279053,0.595370352268219,0.46666669845581055,0.8879629373550415],"occluded":false},{"timestamp":4.16,"box":[0.27916666865348816,0.5925925970077515,0.4755208492279053,0.8703703582286835],"occluded":false},{"timestamp":4.42,"box":[0.29322919249534607,0.5925925970077515,0.48072920739650726,0.8564814925193787],"occluded":false},{"timestamp":4.68,"box":[0.30416667461395264,0.5888888835906982,0.4828125238418579,0.8416666686534882],"occluded":false},{"timestamp":4.94,"box":[0.3020833432674408,0.5888888835906982,0.4802083522081375,0.8407407402992249],"occluded":false},{"timestamp":5.2,"box":[0.2994791865348816,0.5935184955596924,0.47812503576278687,0.8462962806224823],"occluded":false},{"timestamp":5.46,"box":[0.29739585518836975,0.5981481671333313,0.476041704416275,0.8500000238418579],"occluded":false},{"timestamp":5.72,"box":[0.29739585518836975,0.605555534362793,0.47447919845581055,0.8592592477798462],"occluded":false},{"timestamp":5.98,"box":[0.29635417461395264,0.6212962865829468,0.4739583432674408,0.875],"occluded":false},{"timestamp":6.24,"box":[0.2953125238418579,0.635185182094574,0.4734375327825546,0.8898147940635681],"occluded":false},{"timestamp":6.5,"box":[0.30520835518836975,0.6388888955116272,0.47968752682209015,0.8879629671573639],"occluded":false},{"timestamp":6.76,"box":[0.31666669249534607,0.6296296119689941,0.4854166954755783,0.8666666448116302],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.33281251788139343,0.6194444298744202,0.4947916865348816,0.849074050784111],"occluded":false},{"timestamp":7.28,"box":[0.36354169249534607,0.6129629611968994,0.5208333730697632,0.8324074000120163],"occluded":false},{"timestamp":7.54,"box":[0.40000003576278687,0.6037036776542664,0.5500000417232513,0.8111110776662827],"occluded":false},{"timestamp":7.8,"box":[0.43802085518836975,0.5962963104248047,0.5807292014360428,0.7916666716337204],"occluded":false},{"timestamp":8.06,"box":[0.47291669249534607,0.5888888835906982,0.6093750298023224,0.7759259194135666],"occluded":false},{"timestamp":8.32,"box":[0.5041667222976685,0.5907407402992249,0.6369792222976685,0.7675925940275192],"occluded":false},{"timestamp":8.58,"box":[0.5442708730697632,0.5898147821426392,0.6765625476837158,0.7601851522922516],"occluded":false},{"timestamp":8.84,"box":[0.5838541984558105,0.5888888835906982,0.7166666984558105,0.7555555403232574],"occluded":false},{"timestamp":9.1,"box":[0.6192708611488342,0.5916666388511658,0.7567708641290665,0.7564814537763596],"occluded":false},{"timestamp":9.36,"box":[0.6531250476837158,0.6027777791023254,0.7968750596046448,0.7685185223817825],"occluded":false},{"timestamp":9.620000000000001,"box":[0.6817708611488342,0.614814817905426,0.8328125327825546,0.7824074029922485],"occluded":false},{"timestamp":9.88,"box":[0.7078125476837158,0.6268518567085266,0.8671875596046448,0.7999999970197678],"occluded":false},{"timestamp":10.14,"box":[0.7494792342185974,0.6435185074806213,0.9260417371988297,0.8324074000120163],"occluded":false},{"timestamp":10.4,"box":[0.7953125238418579,0.6675925850868225,0.9932291954755783,0.8740740567445755],"occluded":false},{"timestamp":10.66,"box":[0.8447917103767395,0.6953703761100769,0.9994792193174362,0.9027777761220932],"occluded":false},{"timestamp":10.92,"box":[0.9010416865348816,0.7333333492279053,0.9994791895151138,0.9500000178813934],"occluded":false},{"timestamp":11.18,"box":[0.9718750715255737,0.805555522441864,0.9994792398065329,0.9722221791744232],"occluded":false}],"mean_score":0.9179486632347107}],"dropped":{},"fates":[{"noun_phrase":"huge boulder","object_id":0,"frames":38,"mean_score":0.7419355511665344,"mean_area":0.025775463847660385,"outcome":"kept","presentation_id":1,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"huge boulder","object_id":1,"frames":44,"mean_score":0.9179486632347107,"mean_area":0.04375943887795348,"outcome":"kept","presentation_id":5,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"huge boulder","object_id":2,"frames":43,"mean_score":0.9532163739204407,"mean_area":0.1870735838326715,"outcome":"kept","presentation_id":4,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"red door","object_id":0,"frames":40,"mean_score":0.8324607610702515,"mean_area":0.055129836807804235,"outcome":"kept","presentation_id":3,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"red door","object_id":2,"frames":20,"mean_score":0.9139785170555115,"mean_area":0.008223307812488338,"outcome":"kept","presentation_id":2,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null}],"gates":{"min_track_frames":3,"min_mean_score":0.5,"merge_iou":0.8,"max_tracklets":12}}},"render":{"inputs":{"tracklets":5,"as_video":true,"fps":1.0,"max_frames":30},"outputs":{"media":"/home/zzk/data/video_bbox/vlm_gemini3.5_lite/renders/--cif7Nv6BI_chunk_012_scene_010_choices.mp4","media_bytes":6171902,"frames":[{"index":1,"timestamp":0.0,"file":"--cif7Nv6BI_chunk_012_scene_010_choices.mp4"},{"index":2,"timestamp":1.04,"file":"--cif7Nv6BI_chunk_012_scene_010_choices.mp4"},{"index":3,"timestamp":2.08,"file":"--cif7Nv6BI_chunk_012_scene_010_choices.mp4"},{"index":4,"timestamp":2.86,"file":"--cif7Nv6BI_chunk_012_scene_010_choices.mp4"},{"index":5,"timestamp":3.9,"file":"--cif7Nv6BI_chunk_012_scene_010_choices.mp4"},{"index":6,"timestamp":4.94,"file":"--cif7Nv6BI_chunk_012_scene_010_choices.mp4"},{"index":7,"timestamp":5.98,"file":"--cif7Nv6BI_chunk_012_scene_010_choices.mp4"},{"index":8,"timestamp":6.76,"file":"--cif7Nv6BI_chunk_012_scene_010_choices.mp4"},{"index":9,"timestamp":7.8,"file":"--cif7Nv6BI_chunk_012_scene_010_choices.mp4"},{"index":10,"timestamp":8.84,"file":"--cif7Nv6BI_chunk_012_scene_010_choices.mp4"},{"index":11,"timestamp":9.88,"file":"--cif7Nv6BI_chunk_012_scene_010_choices.mp4"},{"index":12,"timestamp":10.66,"file":"--cif7Nv6BI_chunk_012_scene_010_choices.mp4"},{"index":13,"timestamp":11.700000000000001,"file":"--cif7Nv6BI_chunk_012_scene_010_choices.mp4"},{"index":14,"timestamp":12.74,"file":"--cif7Nv6BI_chunk_012_scene_010_choices.mp4"}]}},"select":{"inputs":{"model":"gemini-3.5-flash-lite","temperature":0.0,"as_video":true,"media_bytes":6171902,"media":["--cif7Nv6BI_chunk_012_scene_010_choices.mp4"],"choice_frames":14,"system_prompt":"You are labelling video data for referring multi-object tracking.\n\nYou will see a numbered sequence of frames from one clip, in time order. Each frame has its own number and timestamp printed in the top-left corner. Candidate objects have been detected for you and drawn on the frames as coloured boxes, each labelled with a number. The same object keeps the same number and the same colour on every frame. Below each frame is a strip of enlarged crops of those same numbered boxes, so you can identify a small or distant object.\n\nFor each description you are given, answer two questions:\n\n1. MATCHING_IDS — which of the numbered objects the description refers to. List every number that matches, or an EMPTY LIST if none of them do. An empty list is a correct and expected answer: some descriptions are deliberately about something that is not in this clip. Do not pick a number just to have picked one, and do not pick a number for an object that merely resembles the description.\n\n2. START_FRAME and END_FRAME — the frame numbers between which the described state actually holds. If the description is \"the car that's turning left\", the interval is the frames in which that car is turning, NOT every frame in which the car is visible. Use the frame numbers printed on the frames. START_FRAME must not be greater than END_FRAME. When MATCHING_IDS is empty, give the full range (the first and last frame numbers); it is ignored.\n\n3. CONFIDENCE — how sure you are of the MATCHING_IDS answer, from 0.0 to 1.0. Be honest: a low number costs the example nothing, and an overconfident wrong answer becomes a wrong label.\n\nRules:\n- Only use numbers that are actually drawn on the frames.\n- Never output coordinates, box values, or pixel positions.\n- Judge motion by comparing the object's position across frames, not by guessing from one frame.\n","user_prompt":"The video has 14 frames, numbered 1 to 14. Each frame is one moment of the clip and has its number burned into the top left corner.\n\nAnswer for each of the following descriptions, in the order given:\n1. the huge boulder acting as a roof for the building\n2. the red door of the stone house\n3. the blue car parked in the village square","questions":["the huge boulder acting as a roof for the building","the red door of the stone house","the blue car parked in the village square"]},"outputs":{"raw_response":"{\n  \"answers\": [\n    {\n      \"description_index\": 1,\n      \"matching_ids\": [\n        4\n      ],\n      \"start_frame\": 1,\n      \"end_frame\": 11,\n      \"confidence\": 1.0\n    },\n    {\n      \"description_index\": 2,\n      \"matching_ids\": [\n        3\n      ],\n      \"start_frame\": 1,\n      \"end_frame\": 11,\n      \"confidence\": 1.0\n    },\n    {\n      \"description_index\": 3,\n      \"matching_ids\": [],\n      \"start_frame\": 1,\n      \"end_frame\": 14,\n      \"confidence\": 1.0\n    }\n  ]\n}","raw_answers":[{"description_index":1,"matching_ids":[4],"start_frame":1,"end_frame":11,"confidence":1.0},{"description_index":2,"matching_ids":[3],"start_frame":1,"end_frame":11,"confidence":1.0},{"description_index":3,"matching_ids":[],"start_frame":1,"end_frame":14,"confidence":1.0}],"answers":[{"expression":"the huge boulder acting as a roof for the building","matching_presentation_ids":[4],"start_choice_frame":1,"end_choice_frame":11,"confidence":1.0,"self_reported_confidence":1.0},{"expression":"the red door of the stone house","matching_presentation_ids":[3],"start_choice_frame":1,"end_choice_frame":11,"confidence":1.0,"self_reported_confidence":1.0},{"expression":"the blue car parked in the village square","matching_presentation_ids":[],"start_choice_frame":1,"end_choice_frame":14,"confidence":1.0,"self_reported_confidence":1.0}]}},"verify":{"inputs":{"expressions":3,"tracklets":5,"selections":3,"gates":"interval, count, unknown ids, confidence, then geometry (motion)"},"outputs":{"kept":[{"expression":"the huge boulder acting as a roof for the building","kind":"spatial","is_negative":false,"tracks":[{"timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88],"boxes":[[0.3489583432674408,0.12037036567926407,0.7296875417232513,0.6277777925133705],[0.3500000238418579,0.12222222238779068,0.7328125536441803,0.6268518343567848],[0.33385419845581055,0.11574073880910873,0.7229167222976685,0.6342592462897301],[0.3083333373069763,0.11296296119689941,0.7229166924953461,0.664814829826355],[0.2848958373069763,0.09537036716938019,0.7322916984558105,0.6916666775941849],[0.26093751192092896,0.12037036567926407,0.736979216337204,0.756481446325779],[0.25416669249534607,0.14814814925193787,0.7421875596046448,0.8009259402751923],[0.2562500238418579,0.14999999105930328,0.7427083849906921,0.7999999672174454],[0.2593750059604645,0.13148148357868195,0.7468750178813934,0.781481459736824],[0.2588541805744171,0.0833333283662796,0.7463541924953461,0.7314814478158951],[0.2604166865348816,0.05000000074505806,0.7505208849906921,0.6972222216427326],[0.2588541805744171,0.031481482088565826,0.749479204416275,0.6685185208916664],[0.2614583373069763,0.031481482088565826,0.7463541924953461,0.6425925865769386],[0.27031251788139343,0.031481482088565826,0.7390625476837158,0.6203703656792641],[0.27656251192092896,0.03055555559694767,0.7250000238418579,0.6092592570930719],[0.2864583432674408,0.0416666641831398,0.7098958790302277,0.6064814701676369],[0.30677086114883423,0.069444440305233,0.705729216337204,0.5999999865889549],[0.3192708492279053,0.0962962955236435,0.6989583671092987,0.6018518656492233],[0.32864585518836975,0.11203703284263611,0.6921875476837158,0.5962962806224823],[0.3265625238418579,0.11481481045484543,0.6875000298023224,0.5962962731719017],[0.32343751192092896,0.11759258806705475,0.6843750178813934,0.5999999791383743],[0.3218750059604645,0.1240740716457367,0.6822916865348816,0.6027777791023254],[0.31979167461395264,0.13240741193294525,0.679166704416275,0.6092592626810074],[0.31979167461395264,0.14907407760620117,0.6786458492279053,0.6268518567085266],[0.31979167461395264,0.16388888657093048,0.679166704416275,0.6407407373189926],[0.32864585518836975,0.17777776718139648,0.6817708611488342,0.6472221910953522],[0.3411458432674408,0.18425925076007843,0.6822916865348816,0.6370370239019394],[0.3552083373069763,0.18888889253139496,0.6854166984558105,0.6259259134531021],[0.38697919249534607,0.19629628956317902,0.7072917222976685,0.6203703433275223],[0.4229166805744171,0.20277777314186096,0.7333333492279053,0.6111111044883728],[0.460416704416275,0.21203702688217163,0.760416716337204,0.6064814627170563],[0.4947916865348816,0.2203703671693802,0.7890625298023224,0.599074050784111],[0.526562511920929,0.2342592477798462,0.8161458671092987,0.6009259223937988],[0.5677083730697632,0.24074073135852814,0.861979216337204,0.6018518358469009],[0.6072916984558105,0.23888888955116272,0.9151042103767395,0.6027777791023254],[0.6453125476837158,0.2370370328426361,0.9739584028720856,0.6092592477798462],[0.6791666746139526,0.2361111044883728,0.999479204416275,0.6222222149372101],[0.7098958492279053,0.2361111044883728,0.999479204416275,0.6370370090007782],[0.7364583611488342,0.2361111044883728,0.999479204416275,0.8870370388031006]]}]},{"expression":"the red door of the stone house","kind":"appearance","is_negative":false,"tracks":[{"timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88],"boxes":[[0.5531250238418579,0.7064814567565918,0.610416691750288,0.8157407119870186],[0.5536458492279053,0.7046296000480652,0.6125000193715096,0.8157407119870186],[0.01875000074505806,0.7120370268821716,0.603125024586916,0.9842592477798462],[0.004687500186264515,0.7481481432914734,0.5953125478699803,0.9990740716457367],[0.01875000074505806,0.7787036895751953,0.5979167111217976,0.9990740567445755],[0.07447917014360428,0.8490740656852722,0.5942708924412727,0.9990740567445755],[0.07447917014360428,0.8907407522201538,0.5989583805203438,0.9990740865468979],[0.06041666865348816,0.8907407522201538,0.6000000536441803,0.9990740865468979],[0.5260416865348816,0.8740740418434143,0.6015625223517418,0.9722221866250038],[0.5250000357627869,0.824999988079071,0.6000000387430191,0.970370352268219],[0.5244792103767395,0.7935184836387634,0.5989583805203438,0.9361110776662827],[0.5218750238418579,0.7638888955116272,0.5958333611488342,0.9046296328306198],[0.5234375,0.7342592477798462,0.5963541716337204,0.8731481283903122],[0.5234375,0.7092592716217041,0.5942708402872086,0.8425925970077515],[0.518750011920929,0.6953703761100769,0.5859375149011612,0.8222222328186035],[0.5151041746139526,0.6879629492759705,0.5786458477377892,0.8083333149552345],[0.5218750238418579,0.6814814805984497,0.5822916924953461,0.7944444417953491],[0.5234375,0.6759259104728699,0.5812500044703484,0.784259244799614],[0.02187500149011612,0.6685184836387634,0.5807292237877846,0.9629629254341125],[0.01875000074505806,0.6675925850868225,0.576562512665987,0.9351851642131805],[0.5197917222976685,0.67037034034729,0.5755208916962147,0.7731481194496155],[0.03281250223517418,0.675000011920929,0.5723958872258663,0.9351851940155029],[0.5166667103767395,0.680555522441864,0.5713542141020298,0.7833333015441895],[0.515625,0.6972222328186035,0.5713541693985462,0.800000011920929],[0.004687500186264515,0.7120370268821716,0.5703125121071935,0.9842592477798462],[0.03281250223517418,0.7138888835906982,0.5755208395421505,0.9842592477798462],[0.0598958358168602,0.7037037014961243,0.5770833715796471,0.970370352268219],[0.08750000596046448,0.6925925612449646,0.5828125178813934,0.9425925612449646],[0.12968750298023224,0.6851851940155029,0.6072916835546494,0.9148148149251938],[0.1848958432674408,0.6759259104728699,0.6338542103767395,0.8731481283903122],[0.24114584922790527,0.6666666269302368,0.6619791984558105,0.8592592179775238],[0.2828125059604645,0.6601851582527161,0.6895833611488342,0.859259232878685],[0.33802086114883423,0.6601851582527161,0.7166667282581329,0.7898147851228714],[0.7098958492279053,0.6611111164093018,0.7557291835546494,0.7453703731298447],[0.7515625357627869,0.6629629731178284,0.800000037997961,0.7500000074505806],[0.7937500476837158,0.67037034034729,0.8453125506639481,0.7620370090007782],[0.8364583849906921,0.6861110925674438,0.8937500528991222,0.7833333164453506],[0.8760417103767395,0.7046296000480652,0.9380208812654018,0.8083333000540733],[0.9135417342185974,0.7240740656852722,0.9812500700354576,0.8351851776242256]]}]}],"rejected":{"the blue car parked in the village square":"no_candidates_retrieved"},"rejection_counts":{"no_candidates_retrieved":1},"row":{"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_012_scene_010.mp4","scene_key":"--cif7Nv6BI_chunk_012_scene_010","category":"test_vehicles","height":1080,"width":1920,"fps":50.0,"num_frames":646,"duration":12.92,"rmot_expressions":[{"expression":"the huge boulder acting as a roof for the building","kind":"spatial","is_negative":false,"tracks":[{"timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88],"boxes":[[0.3489583432674408,0.12037036567926407,0.7296875417232513,0.6277777925133705],[0.3500000238418579,0.12222222238779068,0.7328125536441803,0.6268518343567848],[0.33385419845581055,0.11574073880910873,0.7229167222976685,0.6342592462897301],[0.3083333373069763,0.11296296119689941,0.7229166924953461,0.664814829826355],[0.2848958373069763,0.09537036716938019,0.7322916984558105,0.6916666775941849],[0.26093751192092896,0.12037036567926407,0.736979216337204,0.756481446325779],[0.25416669249534607,0.14814814925193787,0.7421875596046448,0.8009259402751923],[0.2562500238418579,0.14999999105930328,0.7427083849906921,0.7999999672174454],[0.2593750059604645,0.13148148357868195,0.7468750178813934,0.781481459736824],[0.2588541805744171,0.0833333283662796,0.7463541924953461,0.7314814478158951],[0.2604166865348816,0.05000000074505806,0.7505208849906921,0.6972222216427326],[0.2588541805744171,0.031481482088565826,0.749479204416275,0.6685185208916664],[0.2614583373069763,0.031481482088565826,0.7463541924953461,0.6425925865769386],[0.27031251788139343,0.031481482088565826,0.7390625476837158,0.6203703656792641],[0.27656251192092896,0.03055555559694767,0.7250000238418579,0.6092592570930719],[0.2864583432674408,0.0416666641831398,0.7098958790302277,0.6064814701676369],[0.30677086114883423,0.069444440305233,0.705729216337204,0.5999999865889549],[0.3192708492279053,0.0962962955236435,0.6989583671092987,0.6018518656492233],[0.32864585518836975,0.11203703284263611,0.6921875476837158,0.5962962806224823],[0.3265625238418579,0.11481481045484543,0.6875000298023224,0.5962962731719017],[0.32343751192092896,0.11759258806705475,0.6843750178813934,0.5999999791383743],[0.3218750059604645,0.1240740716457367,0.6822916865348816,0.6027777791023254],[0.31979167461395264,0.13240741193294525,0.679166704416275,0.6092592626810074],[0.31979167461395264,0.14907407760620117,0.6786458492279053,0.6268518567085266],[0.31979167461395264,0.16388888657093048,0.679166704416275,0.6407407373189926],[0.32864585518836975,0.17777776718139648,0.6817708611488342,0.6472221910953522],[0.3411458432674408,0.18425925076007843,0.6822916865348816,0.6370370239019394],[0.3552083373069763,0.18888889253139496,0.6854166984558105,0.6259259134531021],[0.38697919249534607,0.19629628956317902,0.7072917222976685,0.6203703433275223],[0.4229166805744171,0.20277777314186096,0.7333333492279053,0.6111111044883728],[0.460416704416275,0.21203702688217163,0.760416716337204,0.6064814627170563],[0.4947916865348816,0.2203703671693802,0.7890625298023224,0.599074050784111],[0.526562511920929,0.2342592477798462,0.8161458671092987,0.6009259223937988],[0.5677083730697632,0.24074073135852814,0.861979216337204,0.6018518358469009],[0.6072916984558105,0.23888888955116272,0.9151042103767395,0.6027777791023254],[0.6453125476837158,0.2370370328426361,0.9739584028720856,0.6092592477798462],[0.6791666746139526,0.2361111044883728,0.999479204416275,0.6222222149372101],[0.7098958492279053,0.2361111044883728,0.999479204416275,0.6370370090007782],[0.7364583611488342,0.2361111044883728,0.999479204416275,0.8870370388031006]]}]},{"expression":"the red door of the stone house","kind":"appearance","is_negative":false,"tracks":[{"timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88],"boxes":[[0.5531250238418579,0.7064814567565918,0.610416691750288,0.8157407119870186],[0.5536458492279053,0.7046296000480652,0.6125000193715096,0.8157407119870186],[0.01875000074505806,0.7120370268821716,0.603125024586916,0.9842592477798462],[0.004687500186264515,0.7481481432914734,0.5953125478699803,0.9990740716457367],[0.01875000074505806,0.7787036895751953,0.5979167111217976,0.9990740567445755],[0.07447917014360428,0.8490740656852722,0.5942708924412727,0.9990740567445755],[0.07447917014360428,0.8907407522201538,0.5989583805203438,0.9990740865468979],[0.06041666865348816,0.8907407522201538,0.6000000536441803,0.9990740865468979],[0.5260416865348816,0.8740740418434143,0.6015625223517418,0.9722221866250038],[0.5250000357627869,0.824999988079071,0.6000000387430191,0.970370352268219],[0.5244792103767395,0.7935184836387634,0.5989583805203438,0.9361110776662827],[0.5218750238418579,0.7638888955116272,0.5958333611488342,0.9046296328306198],[0.5234375,0.7342592477798462,0.5963541716337204,0.8731481283903122],[0.5234375,0.7092592716217041,0.5942708402872086,0.8425925970077515],[0.518750011920929,0.6953703761100769,0.5859375149011612,0.8222222328186035],[0.5151041746139526,0.6879629492759705,0.5786458477377892,0.8083333149552345],[0.5218750238418579,0.6814814805984497,0.5822916924953461,0.7944444417953491],[0.5234375,0.6759259104728699,0.5812500044703484,0.784259244799614],[0.02187500149011612,0.6685184836387634,0.5807292237877846,0.9629629254341125],[0.01875000074505806,0.6675925850868225,0.576562512665987,0.9351851642131805],[0.5197917222976685,0.67037034034729,0.5755208916962147,0.7731481194496155],[0.03281250223517418,0.675000011920929,0.5723958872258663,0.9351851940155029],[0.5166667103767395,0.680555522441864,0.5713542141020298,0.7833333015441895],[0.515625,0.6972222328186035,0.5713541693985462,0.800000011920929],[0.004687500186264515,0.7120370268821716,0.5703125121071935,0.9842592477798462],[0.03281250223517418,0.7138888835906982,0.5755208395421505,0.9842592477798462],[0.0598958358168602,0.7037037014961243,0.5770833715796471,0.970370352268219],[0.08750000596046448,0.6925925612449646,0.5828125178813934,0.9425925612449646],[0.12968750298023224,0.6851851940155029,0.6072916835546494,0.9148148149251938],[0.1848958432674408,0.6759259104728699,0.6338542103767395,0.8731481283903122],[0.24114584922790527,0.6666666269302368,0.6619791984558105,0.8592592179775238],[0.2828125059604645,0.6601851582527161,0.6895833611488342,0.859259232878685],[0.33802086114883423,0.6601851582527161,0.7166667282581329,0.7898147851228714],[0.7098958492279053,0.6611111164093018,0.7557291835546494,0.7453703731298447],[0.7515625357627869,0.6629629731178284,0.800000037997961,0.7500000074505806],[0.7937500476837158,0.67037034034729,0.8453125506639481,0.7620370090007782],[0.8364583849906921,0.6861110925674438,0.8937500528991222,0.7833333164453506],[0.8760417103767395,0.7046296000480652,0.9380208812654018,0.8083333000540733],[0.9135417342185974,0.7240740656852722,0.9812500700354576,0.8351851776242256]]}]}],"rmot_expression_count":2,"rmot_negative_count":0,"rmot_motion_count":0}}}}},{"index":12,"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_012_scene_017.mp4","status":"complete","steps":{"decode":{"inputs":{"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_012_scene_017.mp4","sample_fps":4.0},"outputs":{"width":1920,"height":1080,"source_fps":50.0,"duration_seconds":40.2,"frames_kept":155,"blank_frames_dropped":0,"timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88,10.14,10.4,10.66,10.92,11.18,11.44,11.700000000000001,11.96,12.22,12.48,12.74,13.0,13.26,13.52,13.780000000000001,14.040000000000001,14.3,14.56,14.82,15.08,15.34,15.6,15.860000000000001,16.12,16.38,16.64,16.9,17.16,17.42,17.68,17.94,18.2,18.46,18.72,18.98,19.240000000000002,19.5,19.76,20.02,20.28,20.54,20.8,21.06,21.32,21.580000000000002,21.84,22.1,22.36,22.62,22.88,23.14,23.400000000000002,23.66,23.92,24.18,24.44,24.7,24.96,25.22,25.48,25.740000000000002,26.0,26.26,26.52,26.78,27.04,27.3,27.560000000000002,27.82,28.080000000000002,28.34,28.6,28.86,29.12,29.38,29.64,29.900000000000002,30.16,30.42,30.68,30.94,31.2,31.46,31.720000000000002,31.98,32.24,32.5,32.76,33.02,33.28,33.54,33.8,34.06,34.32,34.58,34.84,35.1,35.36,35.62,35.88,36.14,36.4,36.660000000000004,36.92,37.18,37.44,37.7,37.96,38.22,38.480000000000004,38.74,39.0,39.26,39.52,39.78,40.04],"error":""}},"propose":{"inputs":{"model":"gemini-3.5-flash-lite","temperature":0.4,"video_bytes":44088561,"system_prompt":"You are annotating video clips to build training data for referring multi-object tracking: given a video and a natural-language description, a model must output the boxes of every object matching that description, over time.\n\nYour job is to write the descriptions. You do NOT output coordinates, boxes, pixel positions, or any numbers about where things are. A separate detector finds the objects.\n\nFor every clip you produce a list of items. Each item has two texts that must NOT be the same:\n\nEXPRESSION — what a person would say to single out the object or objects. It may describe motion, temporal order, interaction with other objects, and spatial relations. It should read like something a person would actually say.\n  Good: \"the black car that's turning left\"\n  Good: \"the cyclist who overtakes the bus\"\n  Good: \"the two people who walk out of frame on the right\"\n\nNOUN_PHRASE — the same object as a plain, timeless visual category with its visible attributes, and NOTHING else. The detector is prompted with this text and looks at ONE frame at a time, so anything about time, order, or action is useless to it and makes it fail.\n  Allowed: category nouns, colours, materials, clothing, size, type\n  Forbidden: verbs and participles of action (\"turning\", \"walks\", \"is moving\"), temporal words (\"then\", \"starts\", \"after\", \"first\", \"last\"), relative clauses (\"that ...\", \"which ...\", \"who ...\"), ordinals and rank words (\"first\", \"leftmost\", \"nearest\"), and anything over six words.\n  \"the black car that's turning left\" -> \"black car\"\n  \"the cyclist who overtakes the bus\" -> \"cyclist\"\n  \"the two people who walk out of frame on the right\" -> \"person\"\n\nKIND — what makes the expression hard: \"motion\", \"appearance\", \"spatial\", or \"interaction\". Aim for about 35% \"motion\" items: motion is the part a single-frame model cannot already solve, so it is the most valuable, but a list of only motion items is not usable training data either.\n\nEXPECTED_COUNT — how many distinct objects in the clip match the EXPRESSION. Count carefully; this is checked against the detector's answer, and a wrong count discards the item.\n\nIS_NEGATIVE — at most ONE item may be a negative, and only when you also write at least one non-negative item. A negative is an expression that recombines attributes that ARE present in the clip into a whole that is NOT. If the only black car is moving, \"the parked black car\" is a good negative. Its NOUN_PHRASE must still be the category that IS present (\"black car\"), because that is what makes the absence checkable rather than merely unverified. A negative's EXPECTED_COUNT is 0.\n\nRules that discard an item, so follow them exactly:\n- Never write coordinates, box values, pixel positions, or percentages of the frame.\n- Never let a NOUN_PHRASE contain a verb of action, a temporal word, a relative clause, or an ordinal.\n- Never write an EXPRESSION whose object is not visible in the clip, unless it is marked IS_NEGATIVE.\n- Prefer objects that are clearly visible and identifiable over background clutter.\n- Do not describe the camera or the shot; describe objects in the scene.\n","user_prompt":"Watch this clip and write between 1 and 3 items following the rules. Cover different objects rather than restating one object several ways, and include any negative in the same list.\n"},"outputs":{"raw_response":"{\n  \"items\": [\n    {\n      \"expression\": \"the man with a beard and a backpack walking through the dark streets\",\n      \"noun_phrase\": \"bearded man\",\n      \"kind\": \"motion\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the black car parked near the stone wall\",\n      \"noun_phrase\": \"black car\",\n      \"kind\": \"spatial\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the red sports car parked on the cobblestone street\",\n      \"noun_phrase\": \"red sports car\",\n      \"kind\": \"appearance\",\n      \"expected_count\": 0,\n      \"is_negative\": true\n    }\n  ]\n}","raw_items":[{"expression":"the man with a beard and a backpack walking through the dark streets","noun_phrase":"bearded man","kind":"motion","expected_count":1,"is_negative":false},{"expression":"the black car parked near the stone wall","noun_phrase":"black car","kind":"spatial","expected_count":1,"is_negative":false},{"expression":"the red sports car parked on the cobblestone street","noun_phrase":"red sports car","kind":"appearance","expected_count":0,"is_negative":true}],"accepted":[{"expression":"the man with a beard and a backpack walking through the dark streets","noun_phrase":"bearded man","kind":"motion","expected_count":1,"is_negative":false},{"expression":"the black car parked near the stone wall","noun_phrase":"black car","kind":"spatial","expected_count":1,"is_negative":false},{"expression":"the red sports car parked on the cobblestone street","noun_phrase":"red sports car","kind":"appearance","expected_count":0,"is_negative":true}],"gated":[]}},"track":{"inputs":{"tracker":"Sam31MultiplexTracker","model_id":"/mnt/ceph3/ec/jacky/sam3.1","noun_phrases":["bearded man","black car","red sports car"],"expressions_in":3,"frames":155,"sample_fps":4.0,"frame_size":"1920x1080","timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88,10.14,10.4,10.66,10.92,11.18,11.44,11.700000000000001,11.96,12.22,12.48,12.74,13.0,13.26,13.52,13.780000000000001,14.040000000000001,14.3,14.56,14.82,15.08,15.34,15.6,15.860000000000001,16.12,16.38,16.64,16.9,17.16,17.42,17.68,17.94,18.2,18.46,18.72,18.98,19.240000000000002,19.5,19.76,20.02,20.28,20.54,20.8,21.06,21.32,21.580000000000002,21.84,22.1,22.36,22.62,22.88,23.14,23.400000000000002,23.66,23.92,24.18,24.44,24.7,24.96,25.22,25.48,25.740000000000002,26.0,26.26,26.52,26.78,27.04,27.3,27.560000000000002,27.82,28.080000000000002,28.34,28.6,28.86,29.12,29.38,29.64,29.900000000000002,30.16,30.42,30.68,30.94,31.2,31.46,31.720000000000002,31.98,32.24,32.5,32.76,33.02,33.28,33.54,33.8,34.06,34.32,34.58,34.84,35.1,35.36,35.62,35.88,36.14,36.4,36.660000000000004,36.92,37.18,37.44,37.7,37.96,38.22,38.480000000000004,38.74,39.0,39.26,39.52,39.78,40.04],"session":"one session over 155 JPEG frames; then per phrase: reset_session, add_prompt(text) on frame 0, propagate_in_video","passes":3},"outputs":{"raw_tracks":[{"noun_phrase":"bearded man","object_id":0,"observations":[{"timestamp":0.78,"box":[0.6182292103767395,0.4435185194015503,0.8796875476837158,0.9712963104248047],"occluded":false},{"timestamp":1.04,"box":[0.565625011920929,0.4287036955356598,0.8140625208616257,0.9712962806224823],"occluded":false},{"timestamp":1.3,"box":[0.48854169249534607,0.42314815521240234,0.7468750476837158,0.9731481671333313],"occluded":false},{"timestamp":1.56,"box":[0.35989585518836975,0.4842592477798462,0.5484375357627869,0.9712962806224823],"occluded":false},{"timestamp":1.82,"box":[0.2109375149011612,0.4962962865829468,0.4307291954755783,0.9722222089767456],"occluded":false},{"timestamp":2.08,"box":[0.0,0.4277777671813965,0.18958334624767303,0.9712962508201599],"occluded":false},{"timestamp":3.38,"box":[0.04531250149011612,0.49537035822868347,0.2640625163912773,0.9722222089767456],"occluded":false},{"timestamp":3.64,"box":[0.2786458432674408,0.5148147940635681,0.4567708522081375,0.9722221791744232],"occluded":false},{"timestamp":3.9,"box":[0.4958333671092987,0.4175925850868225,0.663541704416275,0.9712962508201599],"occluded":false},{"timestamp":4.16,"box":[0.6239583492279053,0.4287036955356598,0.8765625357627869,0.9712962806224823],"occluded":false},{"timestamp":4.42,"box":[0.8828125596046448,0.42499998211860657,0.9994792342185974,0.9574073851108551],"occluded":false}],"scores":[0.7037037014961243,0.7037037014961243,0.7037037014961243,0.7037037014961243,0.7037037014961243,0.7037037014961243,0.7037037014961243,0.7037037014961243,0.7037037014961243,0.7037037014961243,null]},{"noun_phrase":"black car","object_id":0,"observations":[{"timestamp":0.0,"box":[0.7473958730697632,0.625,0.9994792342185974,0.7999999970197678],"occluded":false},{"timestamp":0.26,"box":[0.7833333611488342,0.6740740537643433,0.999479204416275,0.838888868689537],"occluded":false},{"timestamp":0.52,"box":[0.8265625238418579,0.6879629492759705,0.999479204416275,0.8499999791383743],"occluded":false},{"timestamp":0.78,"box":[0.8255208730697632,0.6925925612449646,0.9994792193174362,0.8555555194616318],"occluded":false},{"timestamp":1.04,"box":[0.7593750357627869,0.6824073791503906,0.9994792193174362,0.9037036746740341],"occluded":false},{"timestamp":1.3,"box":[0.6817708611488342,0.6564814448356628,0.999479204416275,0.8675925582647324],"occluded":false},{"timestamp":1.56,"box":[0.6114583611488342,0.6833333373069763,0.9145833849906921,0.875],"occluded":false},{"timestamp":1.82,"box":[0.5260416865348816,0.7083333134651184,0.994791716337204,0.8796296119689941],"occluded":false},{"timestamp":2.08,"box":[0.4020833671092987,0.7222222089767456,0.6427083760499954,0.8805555403232574],"occluded":false},{"timestamp":2.34,"box":[0.2994791865348816,0.7175925970077515,0.5421875268220901,0.875],"occluded":false},{"timestamp":2.6,"box":[0.2760416865348816,0.7240740656852722,0.5213541984558105,0.8833333253860474],"occluded":false},{"timestamp":2.86,"box":[0.30937501788139343,0.729629635810852,0.5484375357627869,0.881481483578682],"occluded":false},{"timestamp":3.12,"box":[0.4385416805744171,0.7222222089767456,0.6854166984558105,0.8824073970317841],"occluded":false},{"timestamp":3.38,"box":[0.6619791984558105,0.6962962746620178,0.9911458790302277,0.9990740418434143],"occluded":false}],"scores":[0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919]},{"noun_phrase":"black car","object_id":1,"observations":[{"timestamp":2.6,"box":[0.6010416746139526,0.7259259223937988,0.6848958432674408,0.8407407328486443],"occluded":false},{"timestamp":2.86,"box":[0.628125011920929,0.7407407164573669,0.7145833522081375,0.861111082136631],"occluded":false},{"timestamp":3.12,"box":[0.7682291865348816,0.7490740418434143,0.8755208551883698,0.8824073672294617],"occluded":false}],"scores":[0.8476190567016602,0.8476190567016602,0.8476190567016602]}],"tracklets":[{"noun_phrase":"black car","object_id":0,"presentation_id":1,"observations":[{"timestamp":0.0,"box":[0.7473958730697632,0.625,0.9994792342185974,0.7999999970197678],"occluded":false},{"timestamp":0.26,"box":[0.7833333611488342,0.6740740537643433,0.999479204416275,0.838888868689537],"occluded":false},{"timestamp":0.52,"box":[0.8265625238418579,0.6879629492759705,0.999479204416275,0.8499999791383743],"occluded":false},{"timestamp":0.78,"box":[0.8255208730697632,0.6925925612449646,0.9994792193174362,0.8555555194616318],"occluded":false},{"timestamp":1.04,"box":[0.7593750357627869,0.6824073791503906,0.9994792193174362,0.9037036746740341],"occluded":false},{"timestamp":1.3,"box":[0.6817708611488342,0.6564814448356628,0.999479204416275,0.8675925582647324],"occluded":false},{"timestamp":1.56,"box":[0.6114583611488342,0.6833333373069763,0.9145833849906921,0.875],"occluded":false},{"timestamp":1.82,"box":[0.5260416865348816,0.7083333134651184,0.994791716337204,0.8796296119689941],"occluded":false},{"timestamp":2.08,"box":[0.4020833671092987,0.7222222089767456,0.6427083760499954,0.8805555403232574],"occluded":false},{"timestamp":2.34,"box":[0.2994791865348816,0.7175925970077515,0.5421875268220901,0.875],"occluded":false},{"timestamp":2.6,"box":[0.2760416865348816,0.7240740656852722,0.5213541984558105,0.8833333253860474],"occluded":false},{"timestamp":2.86,"box":[0.30937501788139343,0.729629635810852,0.5484375357627869,0.881481483578682],"occluded":false},{"timestamp":3.12,"box":[0.4385416805744171,0.7222222089767456,0.6854166984558105,0.8824073970317841],"occluded":false},{"timestamp":3.38,"box":[0.6619791984558105,0.6962962746620178,0.9911458790302277,0.9990740418434143],"occluded":false}],"mean_score":0.8987342119216919},{"noun_phrase":"black car","object_id":1,"presentation_id":2,"observations":[{"timestamp":2.6,"box":[0.6010416746139526,0.7259259223937988,0.6848958432674408,0.8407407328486443],"occluded":false},{"timestamp":2.86,"box":[0.628125011920929,0.7407407164573669,0.7145833522081375,0.861111082136631],"occluded":false},{"timestamp":3.12,"box":[0.7682291865348816,0.7490740418434143,0.8755208551883698,0.8824073672294617],"occluded":false}],"mean_score":0.8476190567016602},{"noun_phrase":"bearded man","object_id":0,"presentation_id":3,"observations":[{"timestamp":0.78,"box":[0.6182292103767395,0.4435185194015503,0.8796875476837158,0.9712963104248047],"occluded":false},{"timestamp":1.04,"box":[0.565625011920929,0.4287036955356598,0.8140625208616257,0.9712962806224823],"occluded":false},{"timestamp":1.3,"box":[0.48854169249534607,0.42314815521240234,0.7468750476837158,0.9731481671333313],"occluded":false},{"timestamp":1.56,"box":[0.35989585518836975,0.4842592477798462,0.5484375357627869,0.9712962806224823],"occluded":false},{"timestamp":1.82,"box":[0.2109375149011612,0.4962962865829468,0.4307291954755783,0.9722222089767456],"occluded":false},{"timestamp":2.08,"box":[0.0,0.4277777671813965,0.18958334624767303,0.9712962508201599],"occluded":false},{"timestamp":3.38,"box":[0.04531250149011612,0.49537035822868347,0.2640625163912773,0.9722222089767456],"occluded":false},{"timestamp":3.64,"box":[0.2786458432674408,0.5148147940635681,0.4567708522081375,0.9722221791744232],"occluded":false},{"timestamp":3.9,"box":[0.4958333671092987,0.4175925850868225,0.663541704416275,0.9712962508201599],"occluded":false},{"timestamp":4.16,"box":[0.6239583492279053,0.4287036955356598,0.8765625357627869,0.9712962806224823],"occluded":false},{"timestamp":4.42,"box":[0.8828125596046448,0.42499998211860657,0.9994792342185974,0.9574073851108551],"occluded":false}],"mean_score":0.7037037014961243}],"dropped":{},"fates":[{"noun_phrase":"bearded man","object_id":0,"frames":11,"mean_score":0.7037037014961243,"mean_area":0.1083792835631638,"outcome":"kept","presentation_id":3,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"black car","object_id":0,"frames":14,"mean_score":0.8987342119216919,"mean_area":0.04897070203476206,"outcome":"kept","presentation_id":1,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"black car","object_id":1,"frames":3,"mean_score":0.8476190567016602,"mean_area":0.011446759161326647,"outcome":"kept","presentation_id":2,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null}],"gates":{"min_track_frames":3,"min_mean_score":0.5,"merge_iou":0.8,"max_tracklets":12}}},"render":{"inputs":{"tracklets":3,"as_video":true,"fps":1.0,"max_frames":30},"outputs":{"media":"/home/zzk/data/video_bbox/vlm_gemini3.5_lite/renders/--cif7Nv6BI_chunk_012_scene_017_choices.mp4","media_bytes":7862840,"frames":[{"index":1,"timestamp":0.0,"file":"--cif7Nv6BI_chunk_012_scene_017_choices.mp4"},{"index":2,"timestamp":1.3,"file":"--cif7Nv6BI_chunk_012_scene_017_choices.mp4"},{"index":3,"timestamp":2.86,"file":"--cif7Nv6BI_chunk_012_scene_017_choices.mp4"},{"index":4,"timestamp":4.16,"file":"--cif7Nv6BI_chunk_012_scene_017_choices.mp4"},{"index":5,"timestamp":5.46,"file":"--cif7Nv6BI_chunk_012_scene_017_choices.mp4"},{"index":6,"timestamp":7.0200000000000005,"file":"--cif7Nv6BI_chunk_012_scene_017_choices.mp4"},{"index":7,"timestamp":8.32,"file":"--cif7Nv6BI_chunk_012_scene_017_choices.mp4"},{"index":8,"timestamp":9.620000000000001,"file":"--cif7Nv6BI_chunk_012_scene_017_choices.mp4"},{"index":9,"timestamp":10.92,"file":"--cif7Nv6BI_chunk_012_scene_017_choices.mp4"},{"index":10,"timestamp":12.48,"file":"--cif7Nv6BI_chunk_012_scene_017_choices.mp4"},{"index":11,"timestamp":13.780000000000001,"file":"--cif7Nv6BI_chunk_012_scene_017_choices.mp4"},{"index":12,"timestamp":15.08,"file":"--cif7Nv6BI_chunk_012_scene_017_choices.mp4"},{"index":13,"timestamp":16.64,"file":"--cif7Nv6BI_chunk_012_scene_017_choices.mp4"},{"index":14,"timestamp":17.94,"file":"--cif7Nv6BI_chunk_012_scene_017_choices.mp4"},{"index":15,"timestamp":19.240000000000002,"file":"--cif7Nv6BI_chunk_012_scene_017_choices.mp4"},{"index":16,"timestamp":20.8,"file":"--cif7Nv6BI_chunk_012_scene_017_choices.mp4"},{"index":17,"timestamp":22.1,"file":"--cif7Nv6BI_chunk_012_scene_017_choices.mp4"},{"index":18,"timestamp":23.400000000000002,"file":"--cif7Nv6BI_chunk_012_scene_017_choices.mp4"},{"index":19,"timestamp":24.96,"file":"--cif7Nv6BI_chunk_012_scene_017_choices.mp4"},{"index":20,"timestamp":26.26,"file":"--cif7Nv6BI_chunk_012_scene_017_choices.mp4"},{"index":21,"timestamp":27.560000000000002,"file":"--cif7Nv6BI_chunk_012_scene_017_choices.mp4"},{"index":22,"timestamp":29.12,"file":"--cif7Nv6BI_chunk_012_scene_017_choices.mp4"},{"index":23,"timestamp":30.42,"file":"--cif7Nv6BI_chunk_012_scene_017_choices.mp4"},{"index":24,"timestamp":31.720000000000002,"file":"--cif7Nv6BI_chunk_012_scene_017_choices.mp4"},{"index":25,"timestamp":33.02,"file":"--cif7Nv6BI_chunk_012_scene_017_choices.mp4"},{"index":26,"timestamp":34.58,"file":"--cif7Nv6BI_chunk_012_scene_017_choices.mp4"},{"index":27,"timestamp":35.88,"file":"--cif7Nv6BI_chunk_012_scene_017_choices.mp4"},{"index":28,"timestamp":37.18,"file":"--cif7Nv6BI_chunk_012_scene_017_choices.mp4"},{"index":29,"timestamp":38.74,"file":"--cif7Nv6BI_chunk_012_scene_017_choices.mp4"},{"index":30,"timestamp":40.04,"file":"--cif7Nv6BI_chunk_012_scene_017_choices.mp4"}]}},"select":{"inputs":{"model":"gemini-3.5-flash-lite","temperature":0.0,"as_video":true,"media_bytes":7862840,"media":["--cif7Nv6BI_chunk_012_scene_017_choices.mp4"],"choice_frames":30,"system_prompt":"You are labelling video data for referring multi-object tracking.\n\nYou will see a numbered sequence of frames from one clip, in time order. Each frame has its own number and timestamp printed in the top-left corner. Candidate objects have been detected for you and drawn on the frames as coloured boxes, each labelled with a number. The same object keeps the same number and the same colour on every frame. Below each frame is a strip of enlarged crops of those same numbered boxes, so you can identify a small or distant object.\n\nFor each description you are given, answer two questions:\n\n1. MATCHING_IDS — which of the numbered objects the description refers to. List every number that matches, or an EMPTY LIST if none of them do. An empty list is a correct and expected answer: some descriptions are deliberately about something that is not in this clip. Do not pick a number just to have picked one, and do not pick a number for an object that merely resembles the description.\n\n2. START_FRAME and END_FRAME — the frame numbers between which the described state actually holds. If the description is \"the car that's turning left\", the interval is the frames in which that car is turning, NOT every frame in which the car is visible. Use the frame numbers printed on the frames. START_FRAME must not be greater than END_FRAME. When MATCHING_IDS is empty, give the full range (the first and last frame numbers); it is ignored.\n\n3. CONFIDENCE — how sure you are of the MATCHING_IDS answer, from 0.0 to 1.0. Be honest: a low number costs the example nothing, and an overconfident wrong answer becomes a wrong label.\n\nRules:\n- Only use numbers that are actually drawn on the frames.\n- Never output coordinates, box values, or pixel positions.\n- Judge motion by comparing the object's position across frames, not by guessing from one frame.\n","user_prompt":"The video has 30 frames, numbered 1 to 30. Each frame is one moment of the clip and has its number burned into the top left corner.\n\nAnswer for each of the following descriptions, in the order given:\n1. the man with a beard and a backpack walking through the dark streets\n2. the black car parked near the stone wall\n3. the red sports car parked on the cobblestone street","questions":["the man with a beard and a backpack walking through the dark streets","the black car parked near the stone wall","the red sports car parked on the cobblestone street"]},"outputs":{"raw_response":"{\n  \"answers\": [\n    {\n      \"description_index\": 1,\n      \"matching_ids\": [3],\n      \"start_frame\": 2,\n      \"end_frame\": 4,\n      \"confidence\": 0.95\n    },\n    {\n      \"description_index\": 2,\n      \"matching_ids\": [1],\n      \"start_frame\": 1,\n      \"end_frame\": 3,\n      \"confidence\": 0.95\n    },\n    {\n      \"description_index\": 3,\n      \"matching_ids\": [],\n      \"start_frame\": 1,\n      \"end_frame\": 30,\n      \"confidence\": 1.0\n    }\n  ]\n}","raw_answers":[{"description_index":1,"matching_ids":[3],"start_frame":2,"end_frame":4,"confidence":0.95},{"description_index":2,"matching_ids":[1],"start_frame":1,"end_frame":3,"confidence":0.95},{"description_index":3,"matching_ids":[],"start_frame":1,"end_frame":30,"confidence":1.0}],"answers":[{"expression":"the man with a beard and a backpack walking through the dark streets","matching_presentation_ids":[3],"start_choice_frame":2,"end_choice_frame":4,"confidence":0.95,"self_reported_confidence":0.95},{"expression":"the black car parked near the stone wall","matching_presentation_ids":[1],"start_choice_frame":1,"end_choice_frame":3,"confidence":0.95,"self_reported_confidence":0.95},{"expression":"the red sports car parked on the cobblestone street","matching_presentation_ids":[],"start_choice_frame":1,"end_choice_frame":30,"confidence":1.0,"self_reported_confidence":1.0}]}},"verify":{"inputs":{"expressions":3,"tracklets":3,"selections":3,"gates":"interval, count, unknown ids, confidence, then geometry (motion)"},"outputs":{"kept":[{"expression":"the man with a beard and a backpack walking through the dark streets","kind":"motion","is_negative":false,"tracks":[{"timestamps":[1.3,1.56,1.82,2.08,3.38,3.64,3.9,4.16],"boxes":[[0.48854169249534607,0.42314815521240234,0.7468750476837158,0.9731481671333313],[0.35989585518836975,0.4842592477798462,0.5484375357627869,0.9712962806224823],[0.2109375149011612,0.4962962865829468,0.4307291954755783,0.9722222089767456],[0.0,0.4277777671813965,0.18958334624767303,0.9712962508201599],[0.04531250149011612,0.49537035822868347,0.2640625163912773,0.9722222089767456],[0.2786458432674408,0.5148147940635681,0.4567708522081375,0.9722221791744232],[0.4958333671092987,0.4175925850868225,0.663541704416275,0.9712962508201599],[0.6239583492279053,0.4287036955356598,0.8765625357627869,0.9712962806224823]]}]},{"expression":"the black car parked near the stone wall","kind":"spatial","is_negative":false,"tracks":[{"timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86],"boxes":[[0.7473958730697632,0.625,0.9994792342185974,0.7999999970197678],[0.7833333611488342,0.6740740537643433,0.999479204416275,0.838888868689537],[0.8265625238418579,0.6879629492759705,0.999479204416275,0.8499999791383743],[0.8255208730697632,0.6925925612449646,0.9994792193174362,0.8555555194616318],[0.7593750357627869,0.6824073791503906,0.9994792193174362,0.9037036746740341],[0.6817708611488342,0.6564814448356628,0.999479204416275,0.8675925582647324],[0.6114583611488342,0.6833333373069763,0.9145833849906921,0.875],[0.5260416865348816,0.7083333134651184,0.994791716337204,0.8796296119689941],[0.4020833671092987,0.7222222089767456,0.6427083760499954,0.8805555403232574],[0.2994791865348816,0.7175925970077515,0.5421875268220901,0.875],[0.2760416865348816,0.7240740656852722,0.5213541984558105,0.8833333253860474],[0.30937501788139343,0.729629635810852,0.5484375357627869,0.881481483578682]]}]}],"rejected":{"the red sports car parked on the cobblestone street":"no_candidates_retrieved"},"rejection_counts":{"no_candidates_retrieved":1},"row":{"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_012_scene_017.mp4","scene_key":"--cif7Nv6BI_chunk_012_scene_017","category":"test_vehicles","height":1080,"width":1920,"fps":50.0,"num_frames":2010,"duration":40.2,"rmot_expressions":[{"expression":"the man with a beard and a backpack walking through the dark streets","kind":"motion","is_negative":false,"tracks":[{"timestamps":[1.3,1.56,1.82,2.08,3.38,3.64,3.9,4.16],"boxes":[[0.48854169249534607,0.42314815521240234,0.7468750476837158,0.9731481671333313],[0.35989585518836975,0.4842592477798462,0.5484375357627869,0.9712962806224823],[0.2109375149011612,0.4962962865829468,0.4307291954755783,0.9722222089767456],[0.0,0.4277777671813965,0.18958334624767303,0.9712962508201599],[0.04531250149011612,0.49537035822868347,0.2640625163912773,0.9722222089767456],[0.2786458432674408,0.5148147940635681,0.4567708522081375,0.9722221791744232],[0.4958333671092987,0.4175925850868225,0.663541704416275,0.9712962508201599],[0.6239583492279053,0.4287036955356598,0.8765625357627869,0.9712962806224823]]}]},{"expression":"the black car parked near the stone wall","kind":"spatial","is_negative":false,"tracks":[{"timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86],"boxes":[[0.7473958730697632,0.625,0.9994792342185974,0.7999999970197678],[0.7833333611488342,0.6740740537643433,0.999479204416275,0.838888868689537],[0.8265625238418579,0.6879629492759705,0.999479204416275,0.8499999791383743],[0.8255208730697632,0.6925925612449646,0.9994792193174362,0.8555555194616318],[0.7593750357627869,0.6824073791503906,0.9994792193174362,0.9037036746740341],[0.6817708611488342,0.6564814448356628,0.999479204416275,0.8675925582647324],[0.6114583611488342,0.6833333373069763,0.9145833849906921,0.875],[0.5260416865348816,0.7083333134651184,0.994791716337204,0.8796296119689941],[0.4020833671092987,0.7222222089767456,0.6427083760499954,0.8805555403232574],[0.2994791865348816,0.7175925970077515,0.5421875268220901,0.875],[0.2760416865348816,0.7240740656852722,0.5213541984558105,0.8833333253860474],[0.30937501788139343,0.729629635810852,0.5484375357627869,0.881481483578682]]}]}],"rmot_expression_count":2,"rmot_negative_count":0,"rmot_motion_count":1}}}}},{"index":13,"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_000_scene_042.mp4","status":"complete","steps":{"decode":{"inputs":{"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_000_scene_042.mp4","sample_fps":4.0},"outputs":{"width":1920,"height":1080,"source_fps":50.0,"duration_seconds":10.56,"frames_kept":41,"blank_frames_dropped":0,"timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88,10.14,10.4],"error":""}},"propose":{"inputs":{"model":"gemini-3.5-flash-lite","temperature":0.4,"video_bytes":13357762,"system_prompt":"You are annotating video clips to build training data for referring multi-object tracking: given a video and a natural-language description, a model must output the boxes of every object matching that description, over time.\n\nYour job is to write the descriptions. You do NOT output coordinates, boxes, pixel positions, or any numbers about where things are. A separate detector finds the objects.\n\nFor every clip you produce a list of items. Each item has two texts that must NOT be the same:\n\nEXPRESSION — what a person would say to single out the object or objects. It may describe motion, temporal order, interaction with other objects, and spatial relations. It should read like something a person would actually say.\n  Good: \"the black car that's turning left\"\n  Good: \"the cyclist who overtakes the bus\"\n  Good: \"the two people who walk out of frame on the right\"\n\nNOUN_PHRASE — the same object as a plain, timeless visual category with its visible attributes, and NOTHING else. The detector is prompted with this text and looks at ONE frame at a time, so anything about time, order, or action is useless to it and makes it fail.\n  Allowed: category nouns, colours, materials, clothing, size, type\n  Forbidden: verbs and participles of action (\"turning\", \"walks\", \"is moving\"), temporal words (\"then\", \"starts\", \"after\", \"first\", \"last\"), relative clauses (\"that ...\", \"which ...\", \"who ...\"), ordinals and rank words (\"first\", \"leftmost\", \"nearest\"), and anything over six words.\n  \"the black car that's turning left\" -> \"black car\"\n  \"the cyclist who overtakes the bus\" -> \"cyclist\"\n  \"the two people who walk out of frame on the right\" -> \"person\"\n\nKIND — what makes the expression hard: \"motion\", \"appearance\", \"spatial\", or \"interaction\". Aim for about 35% \"motion\" items: motion is the part a single-frame model cannot already solve, so it is the most valuable, but a list of only motion items is not usable training data either.\n\nEXPECTED_COUNT — how many distinct objects in the clip match the EXPRESSION. Count carefully; this is checked against the detector's answer, and a wrong count discards the item.\n\nIS_NEGATIVE — at most ONE item may be a negative, and only when you also write at least one non-negative item. A negative is an expression that recombines attributes that ARE present in the clip into a whole that is NOT. If the only black car is moving, \"the parked black car\" is a good negative. Its NOUN_PHRASE must still be the category that IS present (\"black car\"), because that is what makes the absence checkable rather than merely unverified. A negative's EXPECTED_COUNT is 0.\n\nRules that discard an item, so follow them exactly:\n- Never write coordinates, box values, pixel positions, or percentages of the frame.\n- Never let a NOUN_PHRASE contain a verb of action, a temporal word, a relative clause, or an ordinal.\n- Never write an EXPRESSION whose object is not visible in the clip, unless it is marked IS_NEGATIVE.\n- Prefer objects that are clearly visible and identifiable over background clutter.\n- Do not describe the camera or the shot; describe objects in the scene.\n","user_prompt":"Watch this clip and write between 1 and 3 items following the rules. Cover different objects rather than restating one object several ways, and include any negative in the same list.\n"},"outputs":{"raw_response":"{\n  \"items\": [\n    {\n      \"expression\": \"the white teapot sitting on the table\",\n      \"noun_phrase\": \"white teapot\",\n      \"kind\": \"appearance\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the glasses filled with orange juice\",\n      \"noun_phrase\": \"glass of orange juice\",\n      \"kind\": \"appearance\",\n      \"expected_count\": 3,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the steaming hot bowl of soup\",\n      \"noun_phrase\": \"bowl of soup\",\n      \"kind\": \"appearance\",\n      \"expected_count\": 0,\n      \"is_negative\": true\n    }\n  ]\n}","raw_items":[{"expression":"the white teapot sitting on the table","noun_phrase":"white teapot","kind":"appearance","expected_count":1,"is_negative":false},{"expression":"the glasses filled with orange juice","noun_phrase":"glass of orange juice","kind":"appearance","expected_count":3,"is_negative":false},{"expression":"the steaming hot bowl of soup","noun_phrase":"bowl of soup","kind":"appearance","expected_count":0,"is_negative":true}],"accepted":[{"expression":"the white teapot sitting on the table","noun_phrase":"white teapot","kind":"appearance","expected_count":1,"is_negative":false},{"expression":"the glasses filled with orange juice","noun_phrase":"glass of orange juice","kind":"appearance","expected_count":3,"is_negative":false},{"expression":"the steaming hot bowl of soup","noun_phrase":"bowl of soup","kind":"appearance","expected_count":0,"is_negative":true}],"gated":[]}},"track":{"inputs":{"tracker":"Sam31MultiplexTracker","model_id":"/mnt/ceph3/ec/jacky/sam3.1","noun_phrases":["bowl of soup","glass of orange juice","white teapot"],"expressions_in":3,"frames":41,"sample_fps":4.0,"frame_size":"1920x1080","timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88,10.14,10.4],"session":"one session over 41 JPEG frames; then per phrase: reset_session, add_prompt(text) on frame 0, propagate_in_video","passes":3},"outputs":{"raw_tracks":[{"noun_phrase":"glass of orange juice","object_id":0,"observations":[{"timestamp":0.0,"box":[0.9088541865348816,0.032407406717538834,0.9994791895151138,0.43333331122994423],"occluded":false},{"timestamp":0.26,"box":[0.8494791984558105,0.03055555559694767,0.999479204416275,0.38703701831400394],"occluded":false},{"timestamp":0.52,"box":[0.7963542342185974,0.03055555559694767,0.9234375804662704,0.3425925765186548],"occluded":false},{"timestamp":0.78,"box":[0.7437500357627869,0.03055555559694767,0.8479167073965073,0.29074073769152164],"occluded":false},{"timestamp":1.04,"box":[0.6994792222976685,0.03055555559694767,0.7906250581145287,0.25185185112059116],"occluded":false},{"timestamp":1.3,"box":[0.6484375596046448,0.03055555559694767,0.7322917282581329,0.2592592481523752],"occluded":false},{"timestamp":1.56,"box":[0.6026042103767395,0.03055555559694767,0.6875000447034836,0.2833333406597376],"occluded":false},{"timestamp":1.82,"box":[0.5692708492279053,0.031481482088565826,0.6859375238418579,0.33796296268701553],"occluded":false},{"timestamp":2.08,"box":[0.5473958849906921,0.03055555559694767,0.6812500655651093,0.424074063077569],"occluded":false},{"timestamp":2.34,"box":[0.5276042222976685,0.03055555559694767,0.6781250685453415,0.7685185167938471],"occluded":false},{"timestamp":2.6,"box":[0.510937511920929,0.06759259104728699,0.6994791924953461,0.8796296417713165],"occluded":false},{"timestamp":2.86,"box":[0.5057291984558105,0.10277777910232544,0.713541716337204,0.9629629254341125],"occluded":false},{"timestamp":3.12,"box":[0.5057291984558105,0.13703703880310059,0.6796875447034836,0.6962962746620178],"occluded":false},{"timestamp":3.38,"box":[0.5208333730697632,0.14537036418914795,0.6984375417232513,0.7111110687255859],"occluded":false},{"timestamp":3.64,"box":[0.5479167103767395,0.14999999105930328,0.7255208790302277,0.7148147970438004],"occluded":false},{"timestamp":3.9,"box":[0.6109375357627869,0.14999999105930328,0.7963542193174362,0.7101851850748062],"occluded":false},{"timestamp":4.16,"box":[0.6916667222976685,0.14259259402751923,0.8822917342185974,0.6972222179174423],"occluded":false},{"timestamp":4.42,"box":[0.7593750357627869,0.14814814925193787,0.9645833820104599,0.7000000178813934],"occluded":false},{"timestamp":4.68,"box":[0.7500000596046448,0.1657407432794571,0.9479167312383652,0.7120370417833328],"occluded":false},{"timestamp":4.94,"box":[0.5677083730697632,0.24074073135852814,0.7427083849906921,0.7731481343507767],"occluded":false}],"scores":[0.8324607610702515,0.8324607610702515,0.8324607610702515,0.8324607610702515,0.8324607610702515,0.8324607610702515,0.8324607610702515,0.8324607610702515,0.8324607610702515,0.8324607610702515,0.8324607610702515,0.8324607610702515,0.8324607610702515,0.8324607610702515,0.8324607610702515,0.8324607610702515,0.8324607610702515,0.8324607610702515,0.8324607610702515,0.8324607610702515]},{"noun_phrase":"glass of orange juice","object_id":1,"observations":[{"timestamp":0.0,"box":[0.5614583492279053,0.031481482088565826,0.6317708492279053,0.24074073880910873],"occluded":false},{"timestamp":0.26,"box":[0.5067708492279053,0.031481482088565826,0.9078125357627869,0.2231481447815895],"occluded":false},{"timestamp":0.52,"box":[0.45781251788139343,0.031481482088565826,0.5197916887700558,0.1953703686594963],"occluded":false},{"timestamp":0.78,"box":[0.40000003576278687,0.03055555559694767,0.46510420739650726,0.1583333257585764],"occluded":false},{"timestamp":1.04,"box":[0.3583333492279053,0.03055555559694767,0.4208333492279053,0.13796296156942844],"occluded":false},{"timestamp":1.3,"box":[0.30416667461395264,0.03055555559694767,0.3682291805744171,0.13611111231148243],"occluded":false},{"timestamp":1.56,"box":[0.24895834922790527,0.03055555559694767,0.31562501937150955,0.1611111108213663],"occluded":false},{"timestamp":1.82,"box":[0.19062501192092896,0.03055555559694767,0.2723958492279053,0.21111110784113407],"occluded":false},{"timestamp":2.08,"box":[0.13645833730697632,0.03055555559694767,0.23750001192092896,0.29166666604578495],"occluded":false},{"timestamp":2.34,"box":[0.08750000596046448,0.031481482088565826,0.20468751341104507,0.373148150742054],"occluded":false},{"timestamp":2.6,"box":[0.03593750298023224,0.03055555559694767,0.1744791716337204,0.44537035562098026],"occluded":false},{"timestamp":2.86,"box":[0.00364583358168602,0.03055555559694767,0.15416667982935905,0.4907407257705927],"occluded":false},{"timestamp":3.12,"box":[0.0,0.04722222313284874,0.14791667461395264,0.521296288818121],"occluded":false},{"timestamp":3.38,"box":[0.008333333767950535,0.049074072390794754,0.16041667107492685,0.5268518514931202],"occluded":false},{"timestamp":3.64,"box":[0.0442708358168602,0.05000000074505806,0.18906251341104507,0.5222222097218037],"occluded":false},{"timestamp":3.9,"box":[0.12604168057441711,0.049074072390794754,0.2562500238418579,0.5101851709187031],"occluded":false},{"timestamp":4.16,"box":[0.22552084922790527,0.03703703731298447,0.33802085369825363,0.4833333119750023],"occluded":false},{"timestamp":4.42,"box":[0.3098958432674408,0.037962961941957474,0.408854179084301,0.47685183957219124],"occluded":false},{"timestamp":4.68,"box":[0.30104169249534607,0.06851851940155029,0.40156253427267075,0.5018518567085266],"occluded":false},{"timestamp":4.94,"box":[0.08750000596046448,0.16851851344108582,0.22291667759418488,0.6166666448116302],"occluded":false}],"scores":[0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116]},{"noun_phrase":"glass of orange juice","object_id":2,"observations":[{"timestamp":0.26,"box":[0.9213542342185974,0.11574073880910873,0.999479241669178,0.5787036940455437],"occluded":false},{"timestamp":0.52,"box":[0.8593750596046448,0.06296296417713165,0.9994792342185974,0.5287037044763565],"occluded":false},{"timestamp":0.78,"box":[0.8015625476837158,0.031481482088565826,0.9520833939313889,0.4759259298443794],"occluded":false},{"timestamp":1.04,"box":[0.753125011920929,0.03055555559694767,0.8947916924953461,0.44259257055819035],"occluded":false},{"timestamp":1.3,"box":[0.7026041746139526,0.031481482088565826,0.8359375149011612,0.43888888508081436],"occluded":false},{"timestamp":1.56,"box":[0.6640625596046448,0.031481482088565826,0.7932292371988297,0.4759259298443794],"occluded":false},{"timestamp":1.82,"box":[0.6322916746139526,0.04537037014961243,0.7682291865348816,0.5527777969837189],"occluded":false},{"timestamp":2.08,"box":[0.6161458492279053,0.14814814925193787,0.7562500238418579,0.6703703701496124],"occluded":false},{"timestamp":2.34,"box":[0.5984375476837158,0.25555554032325745,0.7437500506639481,0.7842592298984528],"occluded":false},{"timestamp":2.6,"box":[0.5854166746139526,0.35092592239379883,0.7390625178813934,0.8972222208976746],"occluded":false},{"timestamp":2.86,"box":[0.5802083611488342,0.4148148000240326,0.7463542073965073,0.9712962806224823],"occluded":false},{"timestamp":3.12,"box":[0.5796875357627869,0.46574074029922485,0.7552083730697632,0.9712963104248047],"occluded":false},{"timestamp":3.38,"box":[0.5947917103767395,0.4824073910713196,0.7776042222976685,0.9712962806224823],"occluded":false},{"timestamp":3.64,"box":[0.6250000596046448,0.4842592477798462,0.81145840883255,0.9712962806224823],"occluded":false},{"timestamp":3.9,"box":[0.6947916746139526,0.47870370745658875,0.8921875208616257,0.9712962806224823],"occluded":false},{"timestamp":4.16,"box":[0.7828125357627869,0.4611110985279083,0.9968750476837158,0.9712962806224823],"occluded":false},{"timestamp":4.42,"box":[0.8583333492279053,0.4611110985279083,0.9994791895151138,0.9712962806224823],"occluded":false},{"timestamp":4.68,"box":[0.8479167222976685,0.4722222089767456,0.9994792342185974,0.9712962806224823],"occluded":false},{"timestamp":4.94,"box":[0.6588541865348816,0.5342592597007751,0.8208333551883698,0.9712962806224823],"occluded":false},{"timestamp":5.2,"box":[0.06041666865348816,0.43518516421318054,0.5250000357627869,0.9990740716457367],"occluded":false}],"scores":[0.9101123809814453,0.9101123809814453,0.9101123809814453,0.9101123809814453,0.9101123809814453,0.9101123809814453,0.9101123809814453,0.9101123809814453,0.9101123809814453,0.9101123809814453,0.9101123809814453,0.9101123809814453,0.9101123809814453,0.9101123809814453,0.9101123809814453,0.9101123809814453,0.9101123809814453,0.9101123809814453,0.9101123809814453,0.9101123809814453]},{"noun_phrase":"white teapot","object_id":0,"observations":[{"timestamp":0.0,"box":[0.4442708492279053,0.1953703612089157,0.7598958611488342,0.6703703552484512],"occluded":false},{"timestamp":0.26,"box":[0.38072919845581055,0.1675925850868225,0.8979167342185974,0.6379629373550415],"occluded":false},{"timestamp":0.52,"box":[0.3203125298023224,0.13148148357868195,0.6312500536441803,0.6055555492639542],"occluded":false},{"timestamp":0.78,"box":[0.25416669249534607,0.08703703433275223,0.5713542103767395,0.5712962821125984],"occluded":false},{"timestamp":1.04,"box":[0.19583334028720856,0.060185182839632034,0.5192708522081375,0.5555555410683155],"occluded":false},{"timestamp":1.3,"box":[0.125,0.06388888508081436,0.46041667461395264,0.5749999657273293],"occluded":false},{"timestamp":1.56,"box":[0.04739583656191826,0.1111111119389534,0.6479166857898235,0.6472222283482552],"occluded":false},{"timestamp":1.82,"box":[0.0,0.19907407462596893,0.3526041805744171,0.7759259194135666],"occluded":false},{"timestamp":2.08,"box":[0.0,0.29629629850387573,0.31302085518836975,0.9592592716217041],"occluded":false},{"timestamp":2.34,"box":[0.0,0.49444442987442017,0.27552086114883423,0.9712962806224823],"occluded":false},{"timestamp":2.6,"box":[0.0,0.6472222208976746,0.21562501788139343,0.9712962806224823],"occluded":false},{"timestamp":2.86,"box":[0.0,0.7453703284263611,0.17343750596046448,0.9712962508201599],"occluded":false},{"timestamp":3.12,"box":[0.0,0.824999988079071,0.13125000894069672,0.9712962806224823],"occluded":false},{"timestamp":3.38,"box":[0.0,0.825925886631012,0.13333334028720856,0.9712962508201599],"occluded":false},{"timestamp":3.64,"box":[0.0,0.7944444417953491,0.17395834624767303,0.9712962955236435],"occluded":false},{"timestamp":3.9,"box":[0.0,0.7555555701255798,0.27031251788139343,0.972222238779068],"occluded":false},{"timestamp":4.16,"box":[0.0,0.6092592477798462,0.3828125298023224,0.9722222089767456],"occluded":false},{"timestamp":4.42,"box":[0.07760416716337204,0.5731481313705444,0.4687500223517418,0.9722222089767456],"occluded":false},{"timestamp":4.68,"box":[0.07187500596046448,0.5999999642372131,0.45885419845581055,0.9722221791744232],"occluded":false},{"timestamp":4.94,"box":[0.0,0.8444444537162781,0.18020834028720856,0.9712963104248047],"occluded":false}],"scores":[0.9336099624633789,0.9336099624633789,0.9336099624633789,0.9336099624633789,0.9336099624633789,0.9336099624633789,0.9336099624633789,0.9336099624633789,0.9336099624633789,0.9336099624633789,0.9336099624633789,0.9336099624633789,0.9336099624633789,0.9336099624633789,0.9336099624633789,0.9336099624633789,0.9336099624633789,0.9336099624633789,0.9336099624633789,0.9336099624633789]}],"tracklets":[{"noun_phrase":"glass of orange juice","object_id":0,"presentation_id":1,"observations":[{"timestamp":0.0,"box":[0.9088541865348816,0.032407406717538834,0.9994791895151138,0.43333331122994423],"occluded":false},{"timestamp":0.26,"box":[0.8494791984558105,0.03055555559694767,0.999479204416275,0.38703701831400394],"occluded":false},{"timestamp":0.52,"box":[0.7963542342185974,0.03055555559694767,0.9234375804662704,0.3425925765186548],"occluded":false},{"timestamp":0.78,"box":[0.7437500357627869,0.03055555559694767,0.8479167073965073,0.29074073769152164],"occluded":false},{"timestamp":1.04,"box":[0.6994792222976685,0.03055555559694767,0.7906250581145287,0.25185185112059116],"occluded":false},{"timestamp":1.3,"box":[0.6484375596046448,0.03055555559694767,0.7322917282581329,0.2592592481523752],"occluded":false},{"timestamp":1.56,"box":[0.6026042103767395,0.03055555559694767,0.6875000447034836,0.2833333406597376],"occluded":false},{"timestamp":1.82,"box":[0.5692708492279053,0.031481482088565826,0.6859375238418579,0.33796296268701553],"occluded":false},{"timestamp":2.08,"box":[0.5473958849906921,0.03055555559694767,0.6812500655651093,0.424074063077569],"occluded":false},{"timestamp":2.34,"box":[0.5276042222976685,0.03055555559694767,0.6781250685453415,0.7685185167938471],"occluded":false},{"timestamp":2.6,"box":[0.510937511920929,0.06759259104728699,0.6994791924953461,0.8796296417713165],"occluded":false},{"timestamp":2.86,"box":[0.5057291984558105,0.10277777910232544,0.713541716337204,0.9629629254341125],"occluded":false},{"timestamp":3.12,"box":[0.5057291984558105,0.13703703880310059,0.6796875447034836,0.6962962746620178],"occluded":false},{"timestamp":3.38,"box":[0.5208333730697632,0.14537036418914795,0.6984375417232513,0.7111110687255859],"occluded":false},{"timestamp":3.64,"box":[0.5479167103767395,0.14999999105930328,0.7255208790302277,0.7148147970438004],"occluded":false},{"timestamp":3.9,"box":[0.6109375357627869,0.14999999105930328,0.7963542193174362,0.7101851850748062],"occluded":false},{"timestamp":4.16,"box":[0.6916667222976685,0.14259259402751923,0.8822917342185974,0.6972222179174423],"occluded":false},{"timestamp":4.42,"box":[0.7593750357627869,0.14814814925193787,0.9645833820104599,0.7000000178813934],"occluded":false},{"timestamp":4.68,"box":[0.7500000596046448,0.1657407432794571,0.9479167312383652,0.7120370417833328],"occluded":false},{"timestamp":4.94,"box":[0.5677083730697632,0.24074073135852814,0.7427083849906921,0.7731481343507767],"occluded":false}],"mean_score":0.8324607610702515},{"noun_phrase":"white teapot","object_id":0,"presentation_id":2,"observations":[{"timestamp":0.0,"box":[0.4442708492279053,0.1953703612089157,0.7598958611488342,0.6703703552484512],"occluded":false},{"timestamp":0.26,"box":[0.38072919845581055,0.1675925850868225,0.8979167342185974,0.6379629373550415],"occluded":false},{"timestamp":0.52,"box":[0.3203125298023224,0.13148148357868195,0.6312500536441803,0.6055555492639542],"occluded":false},{"timestamp":0.78,"box":[0.25416669249534607,0.08703703433275223,0.5713542103767395,0.5712962821125984],"occluded":false},{"timestamp":1.04,"box":[0.19583334028720856,0.060185182839632034,0.5192708522081375,0.5555555410683155],"occluded":false},{"timestamp":1.3,"box":[0.125,0.06388888508081436,0.46041667461395264,0.5749999657273293],"occluded":false},{"timestamp":1.56,"box":[0.04739583656191826,0.1111111119389534,0.6479166857898235,0.6472222283482552],"occluded":false},{"timestamp":1.82,"box":[0.0,0.19907407462596893,0.3526041805744171,0.7759259194135666],"occluded":false},{"timestamp":2.08,"box":[0.0,0.29629629850387573,0.31302085518836975,0.9592592716217041],"occluded":false},{"timestamp":2.34,"box":[0.0,0.49444442987442017,0.27552086114883423,0.9712962806224823],"occluded":false},{"timestamp":2.6,"box":[0.0,0.6472222208976746,0.21562501788139343,0.9712962806224823],"occluded":false},{"timestamp":2.86,"box":[0.0,0.7453703284263611,0.17343750596046448,0.9712962508201599],"occluded":false},{"timestamp":3.12,"box":[0.0,0.824999988079071,0.13125000894069672,0.9712962806224823],"occluded":false},{"timestamp":3.38,"box":[0.0,0.825925886631012,0.13333334028720856,0.9712962508201599],"occluded":false},{"timestamp":3.64,"box":[0.0,0.7944444417953491,0.17395834624767303,0.9712962955236435],"occluded":false},{"timestamp":3.9,"box":[0.0,0.7555555701255798,0.27031251788139343,0.972222238779068],"occluded":false},{"timestamp":4.16,"box":[0.0,0.6092592477798462,0.3828125298023224,0.9722222089767456],"occluded":false},{"timestamp":4.42,"box":[0.07760416716337204,0.5731481313705444,0.4687500223517418,0.9722222089767456],"occluded":false},{"timestamp":4.68,"box":[0.07187500596046448,0.5999999642372131,0.45885419845581055,0.9722221791744232],"occluded":false},{"timestamp":4.94,"box":[0.0,0.8444444537162781,0.18020834028720856,0.9712963104248047],"occluded":false}],"mean_score":0.9336099624633789},{"noun_phrase":"glass of orange juice","object_id":1,"presentation_id":3,"observations":[{"timestamp":0.0,"box":[0.5614583492279053,0.031481482088565826,0.6317708492279053,0.24074073880910873],"occluded":false},{"timestamp":0.26,"box":[0.5067708492279053,0.031481482088565826,0.9078125357627869,0.2231481447815895],"occluded":false},{"timestamp":0.52,"box":[0.45781251788139343,0.031481482088565826,0.5197916887700558,0.1953703686594963],"occluded":false},{"timestamp":0.78,"box":[0.40000003576278687,0.03055555559694767,0.46510420739650726,0.1583333257585764],"occluded":false},{"timestamp":1.04,"box":[0.3583333492279053,0.03055555559694767,0.4208333492279053,0.13796296156942844],"occluded":false},{"timestamp":1.3,"box":[0.30416667461395264,0.03055555559694767,0.3682291805744171,0.13611111231148243],"occluded":false},{"timestamp":1.56,"box":[0.24895834922790527,0.03055555559694767,0.31562501937150955,0.1611111108213663],"occluded":false},{"timestamp":1.82,"box":[0.19062501192092896,0.03055555559694767,0.2723958492279053,0.21111110784113407],"occluded":false},{"timestamp":2.08,"box":[0.13645833730697632,0.03055555559694767,0.23750001192092896,0.29166666604578495],"occluded":false},{"timestamp":2.34,"box":[0.08750000596046448,0.031481482088565826,0.20468751341104507,0.373148150742054],"occluded":false},{"timestamp":2.6,"box":[0.03593750298023224,0.03055555559694767,0.1744791716337204,0.44537035562098026],"occluded":false},{"timestamp":2.86,"box":[0.00364583358168602,0.03055555559694767,0.15416667982935905,0.4907407257705927],"occluded":false},{"timestamp":3.12,"box":[0.0,0.04722222313284874,0.14791667461395264,0.521296288818121],"occluded":false},{"timestamp":3.38,"box":[0.008333333767950535,0.049074072390794754,0.16041667107492685,0.5268518514931202],"occluded":false},{"timestamp":3.64,"box":[0.0442708358168602,0.05000000074505806,0.18906251341104507,0.5222222097218037],"occluded":false},{"timestamp":3.9,"box":[0.12604168057441711,0.049074072390794754,0.2562500238418579,0.5101851709187031],"occluded":false},{"timestamp":4.16,"box":[0.22552084922790527,0.03703703731298447,0.33802085369825363,0.4833333119750023],"occluded":false},{"timestamp":4.42,"box":[0.3098958432674408,0.037962961941957474,0.408854179084301,0.47685183957219124],"occluded":false},{"timestamp":4.68,"box":[0.30104169249534607,0.06851851940155029,0.40156253427267075,0.5018518567085266],"occluded":false},{"timestamp":4.94,"box":[0.08750000596046448,0.16851851344108582,0.22291667759418488,0.6166666448116302],"occluded":false}],"mean_score":0.8904109597206116},{"noun_phrase":"glass of orange juice","object_id":2,"presentation_id":4,"observations":[{"timestamp":0.26,"box":[0.9213542342185974,0.11574073880910873,0.999479241669178,0.5787036940455437],"occluded":false},{"timestamp":0.52,"box":[0.8593750596046448,0.06296296417713165,0.9994792342185974,0.5287037044763565],"occluded":false},{"timestamp":0.78,"box":[0.8015625476837158,0.031481482088565826,0.9520833939313889,0.4759259298443794],"occluded":false},{"timestamp":1.04,"box":[0.753125011920929,0.03055555559694767,0.8947916924953461,0.44259257055819035],"occluded":false},{"timestamp":1.3,"box":[0.7026041746139526,0.031481482088565826,0.8359375149011612,0.43888888508081436],"occluded":false},{"timestamp":1.56,"box":[0.6640625596046448,0.031481482088565826,0.7932292371988297,0.4759259298443794],"occluded":false},{"timestamp":1.82,"box":[0.6322916746139526,0.04537037014961243,0.7682291865348816,0.5527777969837189],"occluded":false},{"timestamp":2.08,"box":[0.6161458492279053,0.14814814925193787,0.7562500238418579,0.6703703701496124],"occluded":false},{"timestamp":2.34,"box":[0.5984375476837158,0.25555554032325745,0.7437500506639481,0.7842592298984528],"occluded":false},{"timestamp":2.6,"box":[0.5854166746139526,0.35092592239379883,0.7390625178813934,0.8972222208976746],"occluded":false},{"timestamp":2.86,"box":[0.5802083611488342,0.4148148000240326,0.7463542073965073,0.9712962806224823],"occluded":false},{"timestamp":3.12,"box":[0.5796875357627869,0.46574074029922485,0.7552083730697632,0.9712963104248047],"occluded":false},{"timestamp":3.38,"box":[0.5947917103767395,0.4824073910713196,0.7776042222976685,0.9712962806224823],"occluded":false},{"timestamp":3.64,"box":[0.6250000596046448,0.4842592477798462,0.81145840883255,0.9712962806224823],"occluded":false},{"timestamp":3.9,"box":[0.6947916746139526,0.47870370745658875,0.8921875208616257,0.9712962806224823],"occluded":false},{"timestamp":4.16,"box":[0.7828125357627869,0.4611110985279083,0.9968750476837158,0.9712962806224823],"occluded":false},{"timestamp":4.42,"box":[0.8583333492279053,0.4611110985279083,0.9994791895151138,0.9712962806224823],"occluded":false},{"timestamp":4.68,"box":[0.8479167222976685,0.4722222089767456,0.9994792342185974,0.9712962806224823],"occluded":false},{"timestamp":4.94,"box":[0.6588541865348816,0.5342592597007751,0.8208333551883698,0.9712962806224823],"occluded":false},{"timestamp":5.2,"box":[0.06041666865348816,0.43518516421318054,0.5250000357627869,0.9990740716457367],"occluded":false}],"mean_score":0.9101123809814453}],"dropped":{},"fates":[{"noun_phrase":"glass of orange juice","object_id":0,"frames":20,"mean_score":0.8324607610702515,"mean_area":0.07854752966506331,"outcome":"kept","presentation_id":1,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"glass of orange juice","object_id":1,"frames":20,"mean_score":0.8904109597206116,"mean_area":0.040462192649097874,"outcome":"kept","presentation_id":3,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"glass of orange juice","object_id":2,"frames":20,"mean_score":0.9101123809814453,"mean_area":0.08447835222999624,"outcome":"kept","presentation_id":4,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"white teapot","object_id":0,"frames":20,"mean_score":0.9336099624633789,"mean_area":0.12945153900944723,"outcome":"kept","presentation_id":2,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null}],"gates":{"min_track_frames":3,"min_mean_score":0.5,"merge_iou":0.8,"max_tracklets":12}}},"render":{"inputs":{"tracklets":4,"as_video":true,"fps":1.0,"max_frames":30},"outputs":{"media":"/home/zzk/data/video_bbox/vlm_gemini3.5_lite/renders/--cif7Nv6BI_chunk_000_scene_042_choices.mp4","media_bytes":3109090,"frames":[{"index":1,"timestamp":0.0,"file":"--cif7Nv6BI_chunk_000_scene_042_choices.mp4"},{"index":2,"timestamp":1.04,"file":"--cif7Nv6BI_chunk_000_scene_042_choices.mp4"},{"index":3,"timestamp":2.08,"file":"--cif7Nv6BI_chunk_000_scene_042_choices.mp4"},{"index":4,"timestamp":3.12,"file":"--cif7Nv6BI_chunk_000_scene_042_choices.mp4"},{"index":5,"timestamp":4.16,"file":"--cif7Nv6BI_chunk_000_scene_042_choices.mp4"},{"index":6,"timestamp":5.2,"file":"--cif7Nv6BI_chunk_000_scene_042_choices.mp4"},{"index":7,"timestamp":6.24,"file":"--cif7Nv6BI_chunk_000_scene_042_choices.mp4"},{"index":8,"timestamp":7.28,"file":"--cif7Nv6BI_chunk_000_scene_042_choices.mp4"},{"index":9,"timestamp":8.32,"file":"--cif7Nv6BI_chunk_000_scene_042_choices.mp4"},{"index":10,"timestamp":9.36,"file":"--cif7Nv6BI_chunk_000_scene_042_choices.mp4"},{"index":11,"timestamp":10.4,"file":"--cif7Nv6BI_chunk_000_scene_042_choices.mp4"}]}},"select":{"inputs":{"model":"gemini-3.5-flash-lite","temperature":0.0,"as_video":true,"media_bytes":3109090,"media":["--cif7Nv6BI_chunk_000_scene_042_choices.mp4"],"choice_frames":11,"system_prompt":"You are labelling video data for referring multi-object tracking.\n\nYou will see a numbered sequence of frames from one clip, in time order. Each frame has its own number and timestamp printed in the top-left corner. Candidate objects have been detected for you and drawn on the frames as coloured boxes, each labelled with a number. The same object keeps the same number and the same colour on every frame. Below each frame is a strip of enlarged crops of those same numbered boxes, so you can identify a small or distant object.\n\nFor each description you are given, answer two questions:\n\n1. MATCHING_IDS — which of the numbered objects the description refers to. List every number that matches, or an EMPTY LIST if none of them do. An empty list is a correct and expected answer: some descriptions are deliberately about something that is not in this clip. Do not pick a number just to have picked one, and do not pick a number for an object that merely resembles the description.\n\n2. START_FRAME and END_FRAME — the frame numbers between which the described state actually holds. If the description is \"the car that's turning left\", the interval is the frames in which that car is turning, NOT every frame in which the car is visible. Use the frame numbers printed on the frames. START_FRAME must not be greater than END_FRAME. When MATCHING_IDS is empty, give the full range (the first and last frame numbers); it is ignored.\n\n3. CONFIDENCE — how sure you are of the MATCHING_IDS answer, from 0.0 to 1.0. Be honest: a low number costs the example nothing, and an overconfident wrong answer becomes a wrong label.\n\nRules:\n- Only use numbers that are actually drawn on the frames.\n- Never output coordinates, box values, or pixel positions.\n- Judge motion by comparing the object's position across frames, not by guessing from one frame.\n","user_prompt":"The video has 11 frames, numbered 1 to 11. Each frame is one moment of the clip and has its number burned into the top left corner.\n\nAnswer for each of the following descriptions, in the order given:\n1. the white teapot sitting on the table\n2. the glasses filled with orange juice\n3. the steaming hot bowl of soup","questions":["the white teapot sitting on the table","the glasses filled with orange juice","the steaming hot bowl of soup"]},"outputs":{"raw_response":"{\n  \"answers\": [\n    {\n      \"description_index\": 1,\n      \"matching_ids\": [\n        2\n      ],\n      \"start_frame\": 1,\n      \"end_frame\": 5,\n      \"confidence\": 1.0\n    },\n    {\n      \"description_index\": 2,\n      \"matching_ids\": [\n        1,\n        3,\n        4\n      ],\n      \"start_frame\": 1,\n      \"end_frame\": 6,\n      \"confidence\": 0.95\n    },\n    {\n      \"description_index\": 3,\n      \"matching_ids\": [],\n      \"start_frame\": 1,\n      \"end_frame\": 11,\n      \"confidence\": 1.0\n    }\n  ]\n}","raw_answers":[{"description_index":1,"matching_ids":[2],"start_frame":1,"end_frame":5,"confidence":1.0},{"description_index":2,"matching_ids":[1,3,4],"start_frame":1,"end_frame":6,"confidence":0.95},{"description_index":3,"matching_ids":[],"start_frame":1,"end_frame":11,"confidence":1.0}],"answers":[{"expression":"the white teapot sitting on the table","matching_presentation_ids":[2],"start_choice_frame":1,"end_choice_frame":5,"confidence":1.0,"self_reported_confidence":1.0},{"expression":"the glasses filled with orange juice","matching_presentation_ids":[1,3,4],"start_choice_frame":1,"end_choice_frame":6,"confidence":0.95,"self_reported_confidence":0.95},{"expression":"the steaming hot bowl of soup","matching_presentation_ids":[],"start_choice_frame":1,"end_choice_frame":11,"confidence":1.0,"self_reported_confidence":1.0}]}},"verify":{"inputs":{"expressions":3,"tracklets":4,"selections":3,"gates":"interval, count, unknown ids, confidence, then geometry (motion)"},"outputs":{"kept":[{"expression":"the white teapot sitting on the table","kind":"appearance","is_negative":false,"tracks":[{"timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16],"boxes":[[0.4442708492279053,0.1953703612089157,0.7598958611488342,0.6703703552484512],[0.38072919845581055,0.1675925850868225,0.8979167342185974,0.6379629373550415],[0.3203125298023224,0.13148148357868195,0.6312500536441803,0.6055555492639542],[0.25416669249534607,0.08703703433275223,0.5713542103767395,0.5712962821125984],[0.19583334028720856,0.060185182839632034,0.5192708522081375,0.5555555410683155],[0.125,0.06388888508081436,0.46041667461395264,0.5749999657273293],[0.04739583656191826,0.1111111119389534,0.6479166857898235,0.6472222283482552],[0.0,0.19907407462596893,0.3526041805744171,0.7759259194135666],[0.0,0.29629629850387573,0.31302085518836975,0.9592592716217041],[0.0,0.49444442987442017,0.27552086114883423,0.9712962806224823],[0.0,0.6472222208976746,0.21562501788139343,0.9712962806224823],[0.0,0.7453703284263611,0.17343750596046448,0.9712962508201599],[0.0,0.824999988079071,0.13125000894069672,0.9712962806224823],[0.0,0.825925886631012,0.13333334028720856,0.9712962508201599],[0.0,0.7944444417953491,0.17395834624767303,0.9712962955236435],[0.0,0.7555555701255798,0.27031251788139343,0.972222238779068],[0.0,0.6092592477798462,0.3828125298023224,0.9722222089767456]]}]},{"expression":"the glasses filled with orange juice","kind":"appearance","is_negative":false,"tracks":[{"timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94],"boxes":[[0.9088541865348816,0.032407406717538834,0.9994791895151138,0.43333331122994423],[0.8494791984558105,0.03055555559694767,0.999479204416275,0.38703701831400394],[0.7963542342185974,0.03055555559694767,0.9234375804662704,0.3425925765186548],[0.7437500357627869,0.03055555559694767,0.8479167073965073,0.29074073769152164],[0.6994792222976685,0.03055555559694767,0.7906250581145287,0.25185185112059116],[0.6484375596046448,0.03055555559694767,0.7322917282581329,0.2592592481523752],[0.6026042103767395,0.03055555559694767,0.6875000447034836,0.2833333406597376],[0.5692708492279053,0.031481482088565826,0.6859375238418579,0.33796296268701553],[0.5473958849906921,0.03055555559694767,0.6812500655651093,0.424074063077569],[0.5276042222976685,0.03055555559694767,0.6781250685453415,0.7685185167938471],[0.510937511920929,0.06759259104728699,0.6994791924953461,0.8796296417713165],[0.5057291984558105,0.10277777910232544,0.713541716337204,0.9629629254341125],[0.5057291984558105,0.13703703880310059,0.6796875447034836,0.6962962746620178],[0.5208333730697632,0.14537036418914795,0.6984375417232513,0.7111110687255859],[0.5479167103767395,0.14999999105930328,0.7255208790302277,0.7148147970438004],[0.6109375357627869,0.14999999105930328,0.7963542193174362,0.7101851850748062],[0.6916667222976685,0.14259259402751923,0.8822917342185974,0.6972222179174423],[0.7593750357627869,0.14814814925193787,0.9645833820104599,0.7000000178813934],[0.7500000596046448,0.1657407432794571,0.9479167312383652,0.7120370417833328],[0.5677083730697632,0.24074073135852814,0.7427083849906921,0.7731481343507767]]},{"timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94],"boxes":[[0.5614583492279053,0.031481482088565826,0.6317708492279053,0.24074073880910873],[0.5067708492279053,0.031481482088565826,0.9078125357627869,0.2231481447815895],[0.45781251788139343,0.031481482088565826,0.5197916887700558,0.1953703686594963],[0.40000003576278687,0.03055555559694767,0.46510420739650726,0.1583333257585764],[0.3583333492279053,0.03055555559694767,0.4208333492279053,0.13796296156942844],[0.30416667461395264,0.03055555559694767,0.3682291805744171,0.13611111231148243],[0.24895834922790527,0.03055555559694767,0.31562501937150955,0.1611111108213663],[0.19062501192092896,0.03055555559694767,0.2723958492279053,0.21111110784113407],[0.13645833730697632,0.03055555559694767,0.23750001192092896,0.29166666604578495],[0.08750000596046448,0.031481482088565826,0.20468751341104507,0.373148150742054],[0.03593750298023224,0.03055555559694767,0.1744791716337204,0.44537035562098026],[0.00364583358168602,0.03055555559694767,0.15416667982935905,0.4907407257705927],[0.0,0.04722222313284874,0.14791667461395264,0.521296288818121],[0.008333333767950535,0.049074072390794754,0.16041667107492685,0.5268518514931202],[0.0442708358168602,0.05000000074505806,0.18906251341104507,0.5222222097218037],[0.12604168057441711,0.049074072390794754,0.2562500238418579,0.5101851709187031],[0.22552084922790527,0.03703703731298447,0.33802085369825363,0.4833333119750023],[0.3098958432674408,0.037962961941957474,0.408854179084301,0.47685183957219124],[0.30104169249534607,0.06851851940155029,0.40156253427267075,0.5018518567085266],[0.08750000596046448,0.16851851344108582,0.22291667759418488,0.6166666448116302]]},{"timestamps":[0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2],"boxes":[[0.9213542342185974,0.11574073880910873,0.999479241669178,0.5787036940455437],[0.8593750596046448,0.06296296417713165,0.9994792342185974,0.5287037044763565],[0.8015625476837158,0.031481482088565826,0.9520833939313889,0.4759259298443794],[0.753125011920929,0.03055555559694767,0.8947916924953461,0.44259257055819035],[0.7026041746139526,0.031481482088565826,0.8359375149011612,0.43888888508081436],[0.6640625596046448,0.031481482088565826,0.7932292371988297,0.4759259298443794],[0.6322916746139526,0.04537037014961243,0.7682291865348816,0.5527777969837189],[0.6161458492279053,0.14814814925193787,0.7562500238418579,0.6703703701496124],[0.5984375476837158,0.25555554032325745,0.7437500506639481,0.7842592298984528],[0.5854166746139526,0.35092592239379883,0.7390625178813934,0.8972222208976746],[0.5802083611488342,0.4148148000240326,0.7463542073965073,0.9712962806224823],[0.5796875357627869,0.46574074029922485,0.7552083730697632,0.9712963104248047],[0.5947917103767395,0.4824073910713196,0.7776042222976685,0.9712962806224823],[0.6250000596046448,0.4842592477798462,0.81145840883255,0.9712962806224823],[0.6947916746139526,0.47870370745658875,0.8921875208616257,0.9712962806224823],[0.7828125357627869,0.4611110985279083,0.9968750476837158,0.9712962806224823],[0.8583333492279053,0.4611110985279083,0.9994791895151138,0.9712962806224823],[0.8479167222976685,0.4722222089767456,0.9994792342185974,0.9712962806224823],[0.6588541865348816,0.5342592597007751,0.8208333551883698,0.9712962806224823],[0.06041666865348816,0.43518516421318054,0.5250000357627869,0.9990740716457367]]}]}],"rejected":{"the steaming hot bowl of soup":"no_candidates_retrieved"},"rejection_counts":{"no_candidates_retrieved":1},"row":{"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_000_scene_042.mp4","scene_key":"--cif7Nv6BI_chunk_000_scene_042","category":"test_vehicles","height":1080,"width":1920,"fps":50.0,"num_frames":528,"duration":10.56,"rmot_expressions":[{"expression":"the white teapot sitting on the table","kind":"appearance","is_negative":false,"tracks":[{"timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16],"boxes":[[0.4442708492279053,0.1953703612089157,0.7598958611488342,0.6703703552484512],[0.38072919845581055,0.1675925850868225,0.8979167342185974,0.6379629373550415],[0.3203125298023224,0.13148148357868195,0.6312500536441803,0.6055555492639542],[0.25416669249534607,0.08703703433275223,0.5713542103767395,0.5712962821125984],[0.19583334028720856,0.060185182839632034,0.5192708522081375,0.5555555410683155],[0.125,0.06388888508081436,0.46041667461395264,0.5749999657273293],[0.04739583656191826,0.1111111119389534,0.6479166857898235,0.6472222283482552],[0.0,0.19907407462596893,0.3526041805744171,0.7759259194135666],[0.0,0.29629629850387573,0.31302085518836975,0.9592592716217041],[0.0,0.49444442987442017,0.27552086114883423,0.9712962806224823],[0.0,0.6472222208976746,0.21562501788139343,0.9712962806224823],[0.0,0.7453703284263611,0.17343750596046448,0.9712962508201599],[0.0,0.824999988079071,0.13125000894069672,0.9712962806224823],[0.0,0.825925886631012,0.13333334028720856,0.9712962508201599],[0.0,0.7944444417953491,0.17395834624767303,0.9712962955236435],[0.0,0.7555555701255798,0.27031251788139343,0.972222238779068],[0.0,0.6092592477798462,0.3828125298023224,0.9722222089767456]]}]},{"expression":"the glasses filled with orange juice","kind":"appearance","is_negative":false,"tracks":[{"timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94],"boxes":[[0.9088541865348816,0.032407406717538834,0.9994791895151138,0.43333331122994423],[0.8494791984558105,0.03055555559694767,0.999479204416275,0.38703701831400394],[0.7963542342185974,0.03055555559694767,0.9234375804662704,0.3425925765186548],[0.7437500357627869,0.03055555559694767,0.8479167073965073,0.29074073769152164],[0.6994792222976685,0.03055555559694767,0.7906250581145287,0.25185185112059116],[0.6484375596046448,0.03055555559694767,0.7322917282581329,0.2592592481523752],[0.6026042103767395,0.03055555559694767,0.6875000447034836,0.2833333406597376],[0.5692708492279053,0.031481482088565826,0.6859375238418579,0.33796296268701553],[0.5473958849906921,0.03055555559694767,0.6812500655651093,0.424074063077569],[0.5276042222976685,0.03055555559694767,0.6781250685453415,0.7685185167938471],[0.510937511920929,0.06759259104728699,0.6994791924953461,0.8796296417713165],[0.5057291984558105,0.10277777910232544,0.713541716337204,0.9629629254341125],[0.5057291984558105,0.13703703880310059,0.6796875447034836,0.6962962746620178],[0.5208333730697632,0.14537036418914795,0.6984375417232513,0.7111110687255859],[0.5479167103767395,0.14999999105930328,0.7255208790302277,0.7148147970438004],[0.6109375357627869,0.14999999105930328,0.7963542193174362,0.7101851850748062],[0.6916667222976685,0.14259259402751923,0.8822917342185974,0.6972222179174423],[0.7593750357627869,0.14814814925193787,0.9645833820104599,0.7000000178813934],[0.7500000596046448,0.1657407432794571,0.9479167312383652,0.7120370417833328],[0.5677083730697632,0.24074073135852814,0.7427083849906921,0.7731481343507767]]},{"timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94],"boxes":[[0.5614583492279053,0.031481482088565826,0.6317708492279053,0.24074073880910873],[0.5067708492279053,0.031481482088565826,0.9078125357627869,0.2231481447815895],[0.45781251788139343,0.031481482088565826,0.5197916887700558,0.1953703686594963],[0.40000003576278687,0.03055555559694767,0.46510420739650726,0.1583333257585764],[0.3583333492279053,0.03055555559694767,0.4208333492279053,0.13796296156942844],[0.30416667461395264,0.03055555559694767,0.3682291805744171,0.13611111231148243],[0.24895834922790527,0.03055555559694767,0.31562501937150955,0.1611111108213663],[0.19062501192092896,0.03055555559694767,0.2723958492279053,0.21111110784113407],[0.13645833730697632,0.03055555559694767,0.23750001192092896,0.29166666604578495],[0.08750000596046448,0.031481482088565826,0.20468751341104507,0.373148150742054],[0.03593750298023224,0.03055555559694767,0.1744791716337204,0.44537035562098026],[0.00364583358168602,0.03055555559694767,0.15416667982935905,0.4907407257705927],[0.0,0.04722222313284874,0.14791667461395264,0.521296288818121],[0.008333333767950535,0.049074072390794754,0.16041667107492685,0.5268518514931202],[0.0442708358168602,0.05000000074505806,0.18906251341104507,0.5222222097218037],[0.12604168057441711,0.049074072390794754,0.2562500238418579,0.5101851709187031],[0.22552084922790527,0.03703703731298447,0.33802085369825363,0.4833333119750023],[0.3098958432674408,0.037962961941957474,0.408854179084301,0.47685183957219124],[0.30104169249534607,0.06851851940155029,0.40156253427267075,0.5018518567085266],[0.08750000596046448,0.16851851344108582,0.22291667759418488,0.6166666448116302]]},{"timestamps":[0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2],"boxes":[[0.9213542342185974,0.11574073880910873,0.999479241669178,0.5787036940455437],[0.8593750596046448,0.06296296417713165,0.9994792342185974,0.5287037044763565],[0.8015625476837158,0.031481482088565826,0.9520833939313889,0.4759259298443794],[0.753125011920929,0.03055555559694767,0.8947916924953461,0.44259257055819035],[0.7026041746139526,0.031481482088565826,0.8359375149011612,0.43888888508081436],[0.6640625596046448,0.031481482088565826,0.7932292371988297,0.4759259298443794],[0.6322916746139526,0.04537037014961243,0.7682291865348816,0.5527777969837189],[0.6161458492279053,0.14814814925193787,0.7562500238418579,0.6703703701496124],[0.5984375476837158,0.25555554032325745,0.7437500506639481,0.7842592298984528],[0.5854166746139526,0.35092592239379883,0.7390625178813934,0.8972222208976746],[0.5802083611488342,0.4148148000240326,0.7463542073965073,0.9712962806224823],[0.5796875357627869,0.46574074029922485,0.7552083730697632,0.9712963104248047],[0.5947917103767395,0.4824073910713196,0.7776042222976685,0.9712962806224823],[0.6250000596046448,0.4842592477798462,0.81145840883255,0.9712962806224823],[0.6947916746139526,0.47870370745658875,0.8921875208616257,0.9712962806224823],[0.7828125357627869,0.4611110985279083,0.9968750476837158,0.9712962806224823],[0.8583333492279053,0.4611110985279083,0.9994791895151138,0.9712962806224823],[0.8479167222976685,0.4722222089767456,0.9994792342185974,0.9712962806224823],[0.6588541865348816,0.5342592597007751,0.8208333551883698,0.9712962806224823],[0.06041666865348816,0.43518516421318054,0.5250000357627869,0.9990740716457367]]}]}],"rmot_expression_count":2,"rmot_negative_count":0,"rmot_motion_count":0}}}}},{"index":14,"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_011_scene_019.mp4","status":"complete","steps":{"decode":{"inputs":{"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_011_scene_019.mp4","sample_fps":4.0},"outputs":{"width":1920,"height":1080,"source_fps":50.0,"duration_seconds":11.540000000000001,"frames_kept":45,"blank_frames_dropped":0,"timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88,10.14,10.4,10.66,10.92,11.18,11.44],"error":""}},"propose":{"inputs":{"model":"gemini-3.5-flash-lite","temperature":0.4,"video_bytes":16631425,"system_prompt":"You are annotating video clips to build training data for referring multi-object tracking: given a video and a natural-language description, a model must output the boxes of every object matching that description, over time.\n\nYour job is to write the descriptions. You do NOT output coordinates, boxes, pixel positions, or any numbers about where things are. A separate detector finds the objects.\n\nFor every clip you produce a list of items. Each item has two texts that must NOT be the same:\n\nEXPRESSION — what a person would say to single out the object or objects. It may describe motion, temporal order, interaction with other objects, and spatial relations. It should read like something a person would actually say.\n  Good: \"the black car that's turning left\"\n  Good: \"the cyclist who overtakes the bus\"\n  Good: \"the two people who walk out of frame on the right\"\n\nNOUN_PHRASE — the same object as a plain, timeless visual category with its visible attributes, and NOTHING else. The detector is prompted with this text and looks at ONE frame at a time, so anything about time, order, or action is useless to it and makes it fail.\n  Allowed: category nouns, colours, materials, clothing, size, type\n  Forbidden: verbs and participles of action (\"turning\", \"walks\", \"is moving\"), temporal words (\"then\", \"starts\", \"after\", \"first\", \"last\"), relative clauses (\"that ...\", \"which ...\", \"who ...\"), ordinals and rank words (\"first\", \"leftmost\", \"nearest\"), and anything over six words.\n  \"the black car that's turning left\" -> \"black car\"\n  \"the cyclist who overtakes the bus\" -> \"cyclist\"\n  \"the two people who walk out of frame on the right\" -> \"person\"\n\nKIND — what makes the expression hard: \"motion\", \"appearance\", \"spatial\", or \"interaction\". Aim for about 35% \"motion\" items: motion is the part a single-frame model cannot already solve, so it is the most valuable, but a list of only motion items is not usable training data either.\n\nEXPECTED_COUNT — how many distinct objects in the clip match the EXPRESSION. Count carefully; this is checked against the detector's answer, and a wrong count discards the item.\n\nIS_NEGATIVE — at most ONE item may be a negative, and only when you also write at least one non-negative item. A negative is an expression that recombines attributes that ARE present in the clip into a whole that is NOT. If the only black car is moving, \"the parked black car\" is a good negative. Its NOUN_PHRASE must still be the category that IS present (\"black car\"), because that is what makes the absence checkable rather than merely unverified. A negative's EXPECTED_COUNT is 0.\n\nRules that discard an item, so follow them exactly:\n- Never write coordinates, box values, pixel positions, or percentages of the frame.\n- Never let a NOUN_PHRASE contain a verb of action, a temporal word, a relative clause, or an ordinal.\n- Never write an EXPRESSION whose object is not visible in the clip, unless it is marked IS_NEGATIVE.\n- Prefer objects that are clearly visible and identifiable over background clutter.\n- Do not describe the camera or the shot; describe objects in the scene.\n","user_prompt":"Watch this clip and write between 1 and 3 items following the rules. Cover different objects rather than restating one object several ways, and include any negative in the same list.\n"},"outputs":{"raw_response":"{\n  \"items\": [\n    {\n      \"expression\": \"the stone balcony with the ornate green metal railing\",\n      \"noun_phrase\": \"stone balcony\",\n      \"kind\": \"appearance\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the green wooden double door on the stone building\",\n      \"noun_phrase\": \"green wooden door\",\n      \"kind\": \"appearance\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the white car parked down the cobblestone street\",\n      \"noun_phrase\": \"white car\",\n      \"kind\": \"spatial\",\n      \"expected_count\": 0,\n      \"is_negative\": true\n    }\n  ]\n}","raw_items":[{"expression":"the stone balcony with the ornate green metal railing","noun_phrase":"stone balcony","kind":"appearance","expected_count":1,"is_negative":false},{"expression":"the green wooden double door on the stone building","noun_phrase":"green wooden door","kind":"appearance","expected_count":1,"is_negative":false},{"expression":"the white car parked down the cobblestone street","noun_phrase":"white car","kind":"spatial","expected_count":0,"is_negative":true}],"accepted":[{"expression":"the stone balcony with the ornate green metal railing","noun_phrase":"stone balcony","kind":"appearance","expected_count":1,"is_negative":false},{"expression":"the green wooden double door on the stone building","noun_phrase":"green wooden door","kind":"appearance","expected_count":1,"is_negative":false},{"expression":"the white car parked down the cobblestone street","noun_phrase":"white car","kind":"spatial","expected_count":0,"is_negative":true}],"gated":[]}},"track":{"inputs":{"tracker":"Sam31MultiplexTracker","model_id":"/mnt/ceph3/ec/jacky/sam3.1","noun_phrases":["green wooden door","stone balcony","white car"],"expressions_in":3,"frames":45,"sample_fps":4.0,"frame_size":"1920x1080","timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88,10.14,10.4,10.66,10.92,11.18,11.44],"session":"one session over 45 JPEG frames; then per phrase: reset_session, add_prompt(text) on frame 0, propagate_in_video","passes":3},"outputs":{"raw_tracks":[{"noun_phrase":"green wooden door","object_id":0,"observations":[{"timestamp":0.0,"box":[0.7359375357627869,0.8898147940635681,0.9630208760499954,0.9712962731719017],"occluded":false},{"timestamp":0.26,"box":[0.7088541984558105,0.8768518567085266,0.9281250387430191,0.9712963029742241],"occluded":false},{"timestamp":0.52,"box":[0.6677083969116211,0.8638888597488403,0.8713542371988297,0.9712962657213211],"occluded":false},{"timestamp":0.78,"box":[0.6296875476837158,0.854629635810852,0.8203125596046448,0.9722222238779068],"occluded":false},{"timestamp":1.04,"box":[0.5796875357627869,0.845370352268219,0.7562500387430191,0.9712962806224823],"occluded":false},{"timestamp":1.3,"box":[0.5255208611488342,0.8185185194015503,0.6906250417232513,0.9712962955236435],"occluded":false},{"timestamp":1.56,"box":[0.452604204416275,0.75,0.994791716337204,0.9712962955236435],"occluded":false},{"timestamp":1.82,"box":[0.38229167461395264,0.690740704536438,0.542187511920929,0.9712962508201599],"occluded":false},{"timestamp":2.08,"box":[0.06041666865348816,0.6111111044883728,0.4791666865348816,0.9722222089767456],"occluded":false},{"timestamp":2.34,"box":[0.25468751788139343,0.5249999761581421,0.42760419845581055,0.9722221791744232],"occluded":false},{"timestamp":2.6,"box":[0.20468750596046448,0.4592592418193817,0.6578125357627869,0.9990740716457367],"occluded":false},{"timestamp":2.86,"box":[0.1744791716337204,0.39351850748062134,0.6442708522081375,0.9722222089767456],"occluded":false},{"timestamp":3.12,"box":[0.13229167461395264,0.34166666865348816,0.6031250357627869,0.9712962806224823],"occluded":false},{"timestamp":3.38,"box":[0.06458333879709244,0.31018519401550293,0.5744791999459267,0.9712963104248047],"occluded":false},{"timestamp":3.64,"box":[0.0,0.2888888716697693,0.546875,0.9712962508201599],"occluded":false},{"timestamp":3.9,"box":[0.0,0.28425925970077515,0.5328125357627869,0.9712963104248047],"occluded":false},{"timestamp":4.16,"box":[0.0,0.3037036955356598,0.23541668057441711,0.9712962806224823],"occluded":false},{"timestamp":4.42,"box":[0.04062500223517418,0.3916666507720947,0.5880208872258663,0.970370352268219],"occluded":false},{"timestamp":4.68,"box":[0.1875000149011612,0.5388888716697693,0.38177086412906647,0.9712962806224823],"occluded":false}],"scores":[0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035,0.7852349281311035]},{"noun_phrase":"green wooden door","object_id":1,"observations":[{"timestamp":1.56,"box":[0.9255208969116211,0.8083333373069763,0.9994792342185974,0.9703703671693802],"occluded":false},{"timestamp":1.82,"box":[0.8213542103767395,0.7490740418434143,0.9994792193174362,0.9712962657213211],"occluded":false},{"timestamp":2.08,"box":[0.7395833730697632,0.675000011920929,0.9265625476837158,0.9712963104248047],"occluded":false},{"timestamp":2.34,"box":[0.2552083432674408,0.5249999761581421,0.7411458790302277,0.9712962508201599],"occluded":false},{"timestamp":2.6,"box":[0.6317708492279053,0.5518518686294556,0.74791669100523,0.9157407581806183],"occluded":false},{"timestamp":2.86,"box":[0.6031250357627869,0.49259257316589355,0.7270833775401115,0.8416666388511658],"occluded":false},{"timestamp":3.12,"box":[0.5744792222976685,0.4546296298503876,0.6718750596046448,0.7944444417953491],"occluded":false},{"timestamp":3.38,"box":[0.5427083373069763,0.44814813137054443,0.5911458395421505,0.7916666567325592],"occluded":false},{"timestamp":3.64,"box":[0.515625,0.4694444239139557,0.5593750029802322,0.8120370209217072],"occluded":false},{"timestamp":3.9,"box":[0.499479204416275,0.4870370328426361,0.5406250394880772,0.835185170173645],"occluded":false},{"timestamp":4.16,"box":[0.5010417103767395,0.5009258985519409,0.5427083782851696,0.8462962508201599],"occluded":false}],"scores":[0.8083832263946533,0.8083832263946533,0.8083832263946533,0.8083832263946533,0.8083832263946533,0.8083832263946533,0.8083832263946533,0.8083832263946533,0.8083832263946533,0.8083832263946533,0.8083832263946533]},{"noun_phrase":"green wooden door","object_id":3,"observations":[{"timestamp":2.34,"box":[0.7776042222976685,0.6231481432914734,0.8364583924412727,0.9481481313705444],"occluded":false},{"timestamp":2.6,"box":[0.7250000238418579,0.5685185194015503,0.7729166932404041,0.8694444298744202],"occluded":false},{"timestamp":2.86,"box":[0.6947916746139526,0.5120370388031006,0.7369791753590107,0.7962962985038757],"occluded":false},{"timestamp":3.12,"box":[0.6630208492279053,0.4731481373310089,0.7015625163912773,0.7509258985519409],"occluded":false},{"timestamp":3.38,"box":[0.6286458373069763,0.4731481373310089,0.6614583395421505,0.7444444298744202],"occluded":false},{"timestamp":3.64,"box":[0.6000000238418579,0.5018518567085266,0.627604192122817,0.770370364189148],"occluded":false},{"timestamp":3.9,"box":[0.5838541984558105,0.5240740776062012,0.6088542006909847,0.7935185134410858],"occluded":false},{"timestamp":4.16,"box":[0.5859375596046448,0.5361111164093018,0.6114583946764469,0.8055555522441864],"occluded":false},{"timestamp":4.42,"box":[0.6291667222976685,0.574999988079071,0.6635417230427265,0.8592592477798462],"occluded":false},{"timestamp":4.68,"box":[0.6890625357627869,0.7324073910713196,0.8015625402331352,0.9990740418434143],"occluded":false}],"scores":[0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115]},{"noun_phrase":"green wooden door","object_id":6,"observations":[{"timestamp":9.88,"box":[0.3505208492279053,0.7129629254341125,0.39010418578982353,0.9712962508201599],"occluded":false},{"timestamp":10.14,"box":[0.5041667222976685,0.6083333492279053,0.5395833924412727,0.9231481552124023],"occluded":false},{"timestamp":10.4,"box":[0.5906250476837158,0.5870370268821716,0.627083383500576,0.8962962925434113],"occluded":false},{"timestamp":10.66,"box":[0.6182292103767395,0.5379629731178284,0.6505208797752857,0.8361111283302307],"occluded":false},{"timestamp":10.92,"box":[0.6390625238418579,0.47592592239379883,0.669270858168602,0.7796296179294586],"occluded":false},{"timestamp":11.18,"box":[0.6572917103767395,0.4851851761341095,0.6885417103767395,0.8037036955356598],"occluded":false},{"timestamp":11.44,"box":[0.6755208969116211,0.49444442987442017,0.7078125663101673,0.8287036716938019],"occluded":false}],"scores":[0.7928802371025085,0.7928802371025085,0.7928802371025085,0.7928802371025085,0.7928802371025085,0.7928802371025085,0.7928802371025085]},{"noun_phrase":"stone balcony","object_id":0,"observations":[{"timestamp":0.0,"box":[0.6104167103767395,0.1268518567085266,0.9994792342185974,0.6101851761341095],"occluded":false},{"timestamp":0.26,"box":[0.5973958373069763,0.12870369851589203,0.999479204416275,0.6009259074926376],"occluded":false},{"timestamp":0.52,"box":[0.5770833492279053,0.1518518477678299,0.999479204416275,0.588888868689537],"occluded":false},{"timestamp":0.78,"box":[0.557812511920929,0.1814814805984497,0.999479204416275,0.577777773141861],"occluded":false},{"timestamp":1.04,"box":[0.5239583849906921,0.23055554926395416,0.9994792342185974,0.5805555433034897],"occluded":false},{"timestamp":1.3,"box":[0.48229169845581055,0.21111111342906952,0.9994792342185974,0.5898147970438004],"occluded":false},{"timestamp":1.56,"box":[0.4281250238418579,0.14814814925193787,0.9130208790302277,0.5666666626930237],"occluded":false},{"timestamp":1.82,"box":[0.37187501788139343,0.07222221791744232,0.8312500417232513,0.5259259194135666],"occluded":false},{"timestamp":2.08,"box":[0.31041666865348816,0.03055555559694767,0.7598958611488342,0.4648148212581873],"occluded":false},{"timestamp":2.34,"box":[0.2510416805744171,0.029629629105329514,0.7067708671092987,0.4027777723968029],"occluded":false},{"timestamp":2.6,"box":[0.19427084922790527,0.029629629105329514,0.6671875417232513,0.3481481485068798],"occluded":false},{"timestamp":2.86,"box":[0.15677084028720856,0.029629629105329514,0.6390625387430191,0.2888888828456402],"occluded":false},{"timestamp":3.12,"box":[0.1848958432674408,0.029629629105329514,0.6145833730697632,0.24537036940455437],"occluded":false},{"timestamp":3.38,"box":[0.23906251788139343,0.03055555559694767,0.5843750238418579,0.23518518544733524],"occluded":false},{"timestamp":3.64,"box":[0.23333334922790527,0.03055555559694767,0.5625000298023224,0.2537036929279566],"occluded":false},{"timestamp":3.9,"box":[0.2213541716337204,0.03055555559694767,0.5500000268220901,0.2694444451481104],"occluded":false},{"timestamp":4.16,"box":[0.20520834624767303,0.03055555559694767,0.5510417073965073,0.281481483951211],"occluded":false},{"timestamp":4.42,"box":[0.10000000894069672,0.03055555559694767,0.5843750387430191,0.3259259257465601],"occluded":false},{"timestamp":4.68,"box":[0.19583334028720856,0.031481482088565826,0.6557291895151138,0.5111111178994179],"occluded":false},{"timestamp":4.94,"box":[0.3072916865348816,0.3027777671813965,0.7739583849906921,0.7509258985519409],"occluded":false},{"timestamp":5.2,"box":[0.3604166805744171,0.5064814686775208,0.869791716337204,0.9296296238899231],"occluded":false},{"timestamp":5.46,"box":[0.3843750059604645,0.5370370149612427,0.9770833551883698,0.970370352268219],"occluded":false},{"timestamp":5.72,"box":[0.38697919249534607,0.574999988079071,0.999479204416275,0.9712962806224823],"occluded":false},{"timestamp":5.98,"box":[0.3739583492279053,0.5962963104248047,0.9994792342185974,0.972222238779068],"occluded":false},{"timestamp":6.24,"box":[0.37552085518836975,0.5981481671333313,0.999479204416275,0.972222238779068],"occluded":false},{"timestamp":6.5,"box":[0.3697916865348816,0.5916666388511658,0.9994792342185974,0.9712962508201599],"occluded":false},{"timestamp":6.76,"box":[0.36250001192092896,0.5277777910232544,0.9994792342185974,0.9712963104248047],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.3125000298023224,0.5129629373550415,0.999479204416275,0.9712962508201599],"occluded":false},{"timestamp":7.28,"box":[0.20364584028720856,0.4888888895511627,0.8572917133569717,0.9722222089767456],"occluded":false},{"timestamp":7.54,"box":[0.11041667312383652,0.4037036895751953,0.7546875104308128,0.9712962508201599],"occluded":false},{"timestamp":7.8,"box":[0.05260416865348816,0.32499998807907104,0.6958333551883698,0.9712962508201599],"occluded":false},{"timestamp":8.06,"box":[0.04010416939854622,0.27129629254341125,0.6781250424683094,0.9703703820705414],"occluded":false},{"timestamp":8.32,"box":[0.08906250447034836,0.2796296179294586,0.7046875283122063,0.9694444239139557],"occluded":false},{"timestamp":8.58,"box":[0.18854168057441711,0.38333332538604736,0.8161458671092987,0.970370352268219],"occluded":false},{"timestamp":8.84,"box":[0.3125000298023224,0.4120370149612427,0.9651042520999908,0.970370352268219],"occluded":false},{"timestamp":9.1,"box":[0.41979169845581055,0.29629629850387573,0.9994792342185974,0.8185185194015503],"occluded":false},{"timestamp":9.36,"box":[0.6250000596046448,0.029629629105329514,0.9994792342185974,0.46481479331851006],"occluded":false}],"scores":[0.7460317611694336,0.7460317611694336,0.7460317611694336,0.7460317611694336,0.7460317611694336,0.7460317611694336,0.7460317611694336,0.7460317611694336,0.7460317611694336,0.7460317611694336,0.7460317611694336,0.7460317611694336,0.7460317611694336,0.7460317611694336,0.7460317611694336,0.7460317611694336,0.7460317611694336,0.7460317611694336,0.7460317611694336,0.7460317611694336,0.7460317611694336,0.7460317611694336,0.7460317611694336,0.7460317611694336,0.7460317611694336,0.7460317611694336,0.7460317611694336,0.7460317611694336,0.7460317611694336,0.7460317611694336,0.7460317611694336,0.7460317611694336,0.7460317611694336,0.7460317611694336,0.7460317611694336,0.7460317611694336,0.7460317611694336]},{"noun_phrase":"white car","object_id":0,"observations":[{"timestamp":3.38,"box":[0.8979167342185974,0.5796296000480652,0.9598959051072598,0.6425925642251968],"occluded":false},{"timestamp":3.64,"box":[0.8531250357627869,0.6212962865829468,0.909895870834589,0.6833333224058151],"occluded":false},{"timestamp":3.9,"box":[0.8270833492279053,0.6518518328666687,0.8812500201165676,0.712962944060564],"occluded":false},{"timestamp":4.16,"box":[0.8312500715255737,0.6638888716697693,0.885937575250864,0.7259259074926376],"occluded":false},{"timestamp":4.42,"box":[0.9020833969116211,0.6935185194015503,0.9666667357087135,0.7620370388031006],"occluded":false}],"scores":[0.7460317611694336,0.7460317611694336,0.7460317611694336,0.7460317611694336,0.7460317611694336]},{"noun_phrase":"white car","object_id":1,"observations":[{"timestamp":3.64,"box":[0.9416667222976685,0.6212962865829468,0.9994792267680168,0.698148138821125],"occluded":false},{"timestamp":3.9,"box":[0.9109375476837158,0.654629647731781,0.9859375506639481,0.7287037223577499],"occluded":false},{"timestamp":4.16,"box":[0.9151042103767395,0.6666666269302368,0.9911458790302277,0.7425925508141518],"occluded":false}],"scores":[0.7142857313156128,0.7142857313156128,0.7142857313156128]}],"tracklets":[{"noun_phrase":"green wooden door","object_id":6,"presentation_id":1,"observations":[{"timestamp":9.88,"box":[0.3505208492279053,0.7129629254341125,0.39010418578982353,0.9712962508201599],"occluded":false},{"timestamp":10.14,"box":[0.5041667222976685,0.6083333492279053,0.5395833924412727,0.9231481552124023],"occluded":false},{"timestamp":10.4,"box":[0.5906250476837158,0.5870370268821716,0.627083383500576,0.8962962925434113],"occluded":false},{"timestamp":10.66,"box":[0.6182292103767395,0.5379629731178284,0.6505208797752857,0.8361111283302307],"occluded":false},{"timestamp":10.92,"box":[0.6390625238418579,0.47592592239379883,0.669270858168602,0.7796296179294586],"occluded":false},{"timestamp":11.18,"box":[0.6572917103767395,0.4851851761341095,0.6885417103767395,0.8037036955356598],"occluded":false},{"timestamp":11.44,"box":[0.6755208969116211,0.49444442987442017,0.7078125663101673,0.8287036716938019],"occluded":false}],"mean_score":0.7928802371025085},{"noun_phrase":"green wooden door","object_id":3,"presentation_id":2,"observations":[{"timestamp":2.34,"box":[0.7776042222976685,0.6231481432914734,0.8364583924412727,0.9481481313705444],"occluded":false},{"timestamp":2.6,"box":[0.7250000238418579,0.5685185194015503,0.7729166932404041,0.8694444298744202],"occluded":false},{"timestamp":2.86,"box":[0.6947916746139526,0.5120370388031006,0.7369791753590107,0.7962962985038757],"occluded":false},{"timestamp":3.12,"box":[0.6630208492279053,0.4731481373310089,0.7015625163912773,0.7509258985519409],"occluded":false},{"timestamp":3.38,"box":[0.6286458373069763,0.4731481373310089,0.6614583395421505,0.7444444298744202],"occluded":false},{"timestamp":3.64,"box":[0.6000000238418579,0.5018518567085266,0.627604192122817,0.770370364189148],"occluded":false},{"timestamp":3.9,"box":[0.5838541984558105,0.5240740776062012,0.6088542006909847,0.7935185134410858],"occluded":false},{"timestamp":4.16,"box":[0.5859375596046448,0.5361111164093018,0.6114583946764469,0.8055555522441864],"occluded":false},{"timestamp":4.42,"box":[0.6291667222976685,0.574999988079071,0.6635417230427265,0.8592592477798462],"occluded":false},{"timestamp":4.68,"box":[0.6890625357627869,0.7324073910713196,0.8015625402331352,0.9990740418434143],"occluded":false}],"mean_score":0.9139785170555115},{"noun_phrase":"green wooden door","object_id":1,"presentation_id":3,"observations":[{"timestamp":1.56,"box":[0.9255208969116211,0.8083333373069763,0.9994792342185974,0.9703703671693802],"occluded":false},{"timestamp":1.82,"box":[0.8213542103767395,0.7490740418434143,0.9994792193174362,0.9712962657213211],"occluded":false},{"timestamp":2.08,"box":[0.7395833730697632,0.675000011920929,0.9265625476837158,0.9712963104248047],"occluded":false},{"timestamp":2.34,"box":[0.2552083432674408,0.5249999761581421,0.7411458790302277,0.9712962508201599],"occluded":false},{"timestamp":2.6,"box":[0.6317708492279053,0.5518518686294556,0.74791669100523,0.9157407581806183],"occluded":false},{"timestamp":2.86,"box":[0.6031250357627869,0.49259257316589355,0.7270833775401115,0.8416666388511658],"occluded":false},{"timestamp":3.12,"box":[0.5744792222976685,0.4546296298503876,0.6718750596046448,0.7944444417953491],"occluded":false},{"timestamp":3.38,"box":[0.5427083373069763,0.44814813137054443,0.5911458395421505,0.7916666567325592],"occluded":false},{"timestamp":3.64,"box":[0.515625,0.4694444239139557,0.5593750029802322,0.8120370209217072],"occluded":false},{"timestamp":3.9,"box":[0.499479204416275,0.4870370328426361,0.5406250394880772,0.835185170173645],"occluded":false},{"timestamp":4.16,"box":[0.5010417103767395,0.5009258985519409,0.5427083782851696,0.8462962508201599],"occluded":false}],"mean_score":0.8083832263946533},{"noun_phrase":"white car","object_id":0,"presentation_id":4,"observations":[{"timestamp":3.38,"box":[0.8979167342185974,0.5796296000480652,0.9598959051072598,0.6425925642251968],"occluded":false},{"timestamp":3.64,"box":[0.8531250357627869,0.6212962865829468,0.909895870834589,0.6833333224058151],"occluded":false},{"timestamp":3.9,"box":[0.8270833492279053,0.6518518328666687,0.8812500201165676,0.712962944060564],"occluded":false},{"timestamp":4.16,"box":[0.8312500715255737,0.6638888716697693,0.885937575250864,0.7259259074926376],"occluded":false},{"timestamp":4.42,"box":[0.9020833969116211,0.6935185194015503,0.9666667357087135,0.7620370388031006],"occluded":false}],"mean_score":0.7460317611694336},{"noun_phrase":"stone balcony","object_id":0,"presentation_id":5,"observations":[{"timestamp":0.0,"box":[0.6104167103767395,0.1268518567085266,0.9994792342185974,0.6101851761341095],"occluded":false},{"timestamp":0.26,"box":[0.5973958373069763,0.12870369851589203,0.999479204416275,0.6009259074926376],"occluded":false},{"timestamp":0.52,"box":[0.5770833492279053,0.1518518477678299,0.999479204416275,0.588888868689537],"occluded":false},{"timestamp":0.78,"box":[0.557812511920929,0.1814814805984497,0.999479204416275,0.577777773141861],"occluded":false},{"timestamp":1.04,"box":[0.5239583849906921,0.23055554926395416,0.9994792342185974,0.5805555433034897],"occluded":false},{"timestamp":1.3,"box":[0.48229169845581055,0.21111111342906952,0.9994792342185974,0.5898147970438004],"occluded":false},{"timestamp":1.56,"box":[0.4281250238418579,0.14814814925193787,0.9130208790302277,0.5666666626930237],"occluded":false},{"timestamp":1.82,"box":[0.37187501788139343,0.07222221791744232,0.8312500417232513,0.5259259194135666],"occluded":false},{"timestamp":2.08,"box":[0.31041666865348816,0.03055555559694767,0.7598958611488342,0.4648148212581873],"occluded":false},{"timestamp":2.34,"box":[0.2510416805744171,0.029629629105329514,0.7067708671092987,0.4027777723968029],"occluded":false},{"timestamp":2.6,"box":[0.19427084922790527,0.029629629105329514,0.6671875417232513,0.3481481485068798],"occluded":false},{"timestamp":2.86,"box":[0.15677084028720856,0.029629629105329514,0.6390625387430191,0.2888888828456402],"occluded":false},{"timestamp":3.12,"box":[0.1848958432674408,0.029629629105329514,0.6145833730697632,0.24537036940455437],"occluded":false},{"timestamp":3.38,"box":[0.23906251788139343,0.03055555559694767,0.5843750238418579,0.23518518544733524],"occluded":false},{"timestamp":3.64,"box":[0.23333334922790527,0.03055555559694767,0.5625000298023224,0.2537036929279566],"occluded":false},{"timestamp":3.9,"box":[0.2213541716337204,0.03055555559694767,0.5500000268220901,0.2694444451481104],"occluded":false},{"timestamp":4.16,"box":[0.20520834624767303,0.03055555559694767,0.5510417073965073,0.281481483951211],"occluded":false},{"timestamp":4.42,"box":[0.10000000894069672,0.03055555559694767,0.5843750387430191,0.3259259257465601],"occluded":false},{"timestamp":4.68,"box":[0.19583334028720856,0.031481482088565826,0.6557291895151138,0.5111111178994179],"occluded":false},{"timestamp":4.94,"box":[0.3072916865348816,0.3027777671813965,0.7739583849906921,0.7509258985519409],"occluded":false},{"timestamp":5.2,"box":[0.3604166805744171,0.5064814686775208,0.869791716337204,0.9296296238899231],"occluded":false},{"timestamp":5.46,"box":[0.3843750059604645,0.5370370149612427,0.9770833551883698,0.970370352268219],"occluded":false},{"timestamp":5.72,"box":[0.38697919249534607,0.574999988079071,0.999479204416275,0.9712962806224823],"occluded":false},{"timestamp":5.98,"box":[0.3739583492279053,0.5962963104248047,0.9994792342185974,0.972222238779068],"occluded":false},{"timestamp":6.24,"box":[0.37552085518836975,0.5981481671333313,0.999479204416275,0.972222238779068],"occluded":false},{"timestamp":6.5,"box":[0.3697916865348816,0.5916666388511658,0.9994792342185974,0.9712962508201599],"occluded":false},{"timestamp":6.76,"box":[0.36250001192092896,0.5277777910232544,0.9994792342185974,0.9712963104248047],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.3125000298023224,0.5129629373550415,0.999479204416275,0.9712962508201599],"occluded":false},{"timestamp":7.28,"box":[0.20364584028720856,0.4888888895511627,0.8572917133569717,0.9722222089767456],"occluded":false},{"timestamp":7.54,"box":[0.11041667312383652,0.4037036895751953,0.7546875104308128,0.9712962508201599],"occluded":false},{"timestamp":7.8,"box":[0.05260416865348816,0.32499998807907104,0.6958333551883698,0.9712962508201599],"occluded":false},{"timestamp":8.06,"box":[0.04010416939854622,0.27129629254341125,0.6781250424683094,0.9703703820705414],"occluded":false},{"timestamp":8.32,"box":[0.08906250447034836,0.2796296179294586,0.7046875283122063,0.9694444239139557],"occluded":false},{"timestamp":8.58,"box":[0.18854168057441711,0.38333332538604736,0.8161458671092987,0.970370352268219],"occluded":false},{"timestamp":8.84,"box":[0.3125000298023224,0.4120370149612427,0.9651042520999908,0.970370352268219],"occluded":false},{"timestamp":9.1,"box":[0.41979169845581055,0.29629629850387573,0.9994792342185974,0.8185185194015503],"occluded":false},{"timestamp":9.36,"box":[0.6250000596046448,0.029629629105329514,0.9994792342185974,0.46481479331851006],"occluded":false}],"mean_score":0.7460317611694336},{"noun_phrase":"green wooden door","object_id":0,"presentation_id":6,"observations":[{"timestamp":0.0,"box":[0.7359375357627869,0.8898147940635681,0.9630208760499954,0.9712962731719017],"occluded":false},{"timestamp":0.26,"box":[0.7088541984558105,0.8768518567085266,0.9281250387430191,0.9712963029742241],"occluded":false},{"timestamp":0.52,"box":[0.6677083969116211,0.8638888597488403,0.8713542371988297,0.9712962657213211],"occluded":false},{"timestamp":0.78,"box":[0.6296875476837158,0.854629635810852,0.8203125596046448,0.9722222238779068],"occluded":false},{"timestamp":1.04,"box":[0.5796875357627869,0.845370352268219,0.7562500387430191,0.9712962806224823],"occluded":false},{"timestamp":1.3,"box":[0.5255208611488342,0.8185185194015503,0.6906250417232513,0.9712962955236435],"occluded":false},{"timestamp":1.56,"box":[0.452604204416275,0.75,0.994791716337204,0.9712962955236435],"occluded":false},{"timestamp":1.82,"box":[0.38229167461395264,0.690740704536438,0.542187511920929,0.9712962508201599],"occluded":false},{"timestamp":2.08,"box":[0.06041666865348816,0.6111111044883728,0.4791666865348816,0.9722222089767456],"occluded":false},{"timestamp":2.34,"box":[0.25468751788139343,0.5249999761581421,0.42760419845581055,0.9722221791744232],"occluded":false},{"timestamp":2.6,"box":[0.20468750596046448,0.4592592418193817,0.6578125357627869,0.9990740716457367],"occluded":false},{"timestamp":2.86,"box":[0.1744791716337204,0.39351850748062134,0.6442708522081375,0.9722222089767456],"occluded":false},{"timestamp":3.12,"box":[0.13229167461395264,0.34166666865348816,0.6031250357627869,0.9712962806224823],"occluded":false},{"timestamp":3.38,"box":[0.06458333879709244,0.31018519401550293,0.5744791999459267,0.9712963104248047],"occluded":false},{"timestamp":3.64,"box":[0.0,0.2888888716697693,0.546875,0.9712962508201599],"occluded":false},{"timestamp":3.9,"box":[0.0,0.28425925970077515,0.5328125357627869,0.9712963104248047],"occluded":false},{"timestamp":4.16,"box":[0.0,0.3037036955356598,0.23541668057441711,0.9712962806224823],"occluded":false},{"timestamp":4.42,"box":[0.04062500223517418,0.3916666507720947,0.5880208872258663,0.970370352268219],"occluded":false},{"timestamp":4.68,"box":[0.1875000149011612,0.5388888716697693,0.38177086412906647,0.9712962806224823],"occluded":false}],"mean_score":0.7852349281311035},{"noun_phrase":"white car","object_id":1,"presentation_id":7,"observations":[{"timestamp":3.64,"box":[0.9416667222976685,0.6212962865829468,0.9994792267680168,0.698148138821125],"occluded":false},{"timestamp":3.9,"box":[0.9109375476837158,0.654629647731781,0.9859375506639481,0.7287037223577499],"occluded":false},{"timestamp":4.16,"box":[0.9151042103767395,0.6666666269302368,0.9911458790302277,0.7425925508141518],"occluded":false}],"mean_score":0.7142857313156128}],"dropped":{},"fates":[{"noun_phrase":"green wooden door","object_id":0,"frames":19,"mean_score":0.7852349281311035,"mean_area":0.15639856633993926,"outcome":"kept","presentation_id":6,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"green wooden door","object_id":1,"frames":11,"mean_score":0.8083832263946533,"mean_area":0.045710448195653924,"outcome":"kept","presentation_id":3,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"green wooden door","object_id":3,"frames":10,"mean_score":0.9139785170555115,"mean_area":0.012594328815600363,"outcome":"kept","presentation_id":2,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"green wooden door","object_id":6,"frames":7,"mean_score":0.7928802371025085,"mean_area":0.010314291765456729,"outcome":"kept","presentation_id":1,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"stone balcony","object_id":0,"frames":37,"mean_score":0.7460317611694336,"mean_area":0.2246923043652914,"outcome":"kept","presentation_id":5,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"white car","object_id":0,"frames":5,"mean_score":0.7460317611694336,"mean_area":0.003710455494633888,"outcome":"kept","presentation_id":4,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"white car","object_id":1,"frames":3,"mean_score":0.7142857313156128,"mean_area":0.005257362604990157,"outcome":"kept","presentation_id":7,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null}],"gates":{"min_track_frames":3,"min_mean_score":0.5,"merge_iou":0.8,"max_tracklets":12}}},"render":{"inputs":{"tracklets":7,"as_video":true,"fps":1.0,"max_frames":30},"outputs":{"media":"/home/zzk/data/video_bbox/vlm_gemini3.5_lite/renders/--cif7Nv6BI_chunk_011_scene_019_choices.mp4","media_bytes":3862365,"frames":[{"index":1,"timestamp":0.0,"file":"--cif7Nv6BI_chunk_011_scene_019_choices.mp4"},{"index":2,"timestamp":1.04,"file":"--cif7Nv6BI_chunk_011_scene_019_choices.mp4"},{"index":3,"timestamp":2.08,"file":"--cif7Nv6BI_chunk_011_scene_019_choices.mp4"},{"index":4,"timestamp":3.12,"file":"--cif7Nv6BI_chunk_011_scene_019_choices.mp4"},{"index":5,"timestamp":4.16,"file":"--cif7Nv6BI_chunk_011_scene_019_choices.mp4"},{"index":6,"timestamp":5.2,"file":"--cif7Nv6BI_chunk_011_scene_019_choices.mp4"},{"index":7,"timestamp":6.24,"file":"--cif7Nv6BI_chunk_011_scene_019_choices.mp4"},{"index":8,"timestamp":7.28,"file":"--cif7Nv6BI_chunk_011_scene_019_choices.mp4"},{"index":9,"timestamp":8.32,"file":"--cif7Nv6BI_chunk_011_scene_019_choices.mp4"},{"index":10,"timestamp":9.36,"file":"--cif7Nv6BI_chunk_011_scene_019_choices.mp4"},{"index":11,"timestamp":10.4,"file":"--cif7Nv6BI_chunk_011_scene_019_choices.mp4"},{"index":12,"timestamp":11.44,"file":"--cif7Nv6BI_chunk_011_scene_019_choices.mp4"}]}},"select":{"inputs":{"model":"gemini-3.5-flash-lite","temperature":0.0,"as_video":true,"media_bytes":3862365,"media":["--cif7Nv6BI_chunk_011_scene_019_choices.mp4"],"choice_frames":12,"system_prompt":"You are labelling video data for referring multi-object tracking.\n\nYou will see a numbered sequence of frames from one clip, in time order. Each frame has its own number and timestamp printed in the top-left corner. Candidate objects have been detected for you and drawn on the frames as coloured boxes, each labelled with a number. The same object keeps the same number and the same colour on every frame. Below each frame is a strip of enlarged crops of those same numbered boxes, so you can identify a small or distant object.\n\nFor each description you are given, answer two questions:\n\n1. MATCHING_IDS — which of the numbered objects the description refers to. List every number that matches, or an EMPTY LIST if none of them do. An empty list is a correct and expected answer: some descriptions are deliberately about something that is not in this clip. Do not pick a number just to have picked one, and do not pick a number for an object that merely resembles the description.\n\n2. START_FRAME and END_FRAME — the frame numbers between which the described state actually holds. If the description is \"the car that's turning left\", the interval is the frames in which that car is turning, NOT every frame in which the car is visible. Use the frame numbers printed on the frames. START_FRAME must not be greater than END_FRAME. When MATCHING_IDS is empty, give the full range (the first and last frame numbers); it is ignored.\n\n3. CONFIDENCE — how sure you are of the MATCHING_IDS answer, from 0.0 to 1.0. Be honest: a low number costs the example nothing, and an overconfident wrong answer becomes a wrong label.\n\nRules:\n- Only use numbers that are actually drawn on the frames.\n- Never output coordinates, box values, or pixel positions.\n- Judge motion by comparing the object's position across frames, not by guessing from one frame.\n","user_prompt":"The video has 12 frames, numbered 1 to 12. Each frame is one moment of the clip and has its number burned into the top left corner.\n\nAnswer for each of the following descriptions, in the order given:\n1. the stone balcony with the ornate green metal railing\n2. the green wooden double door on the stone building\n3. the white car parked down the cobblestone street","questions":["the stone balcony with the ornate green metal railing","the green wooden double door on the stone building","the white car parked down the cobblestone street"]},"outputs":{"raw_response":"{\n  \"answers\": [\n    {\n      \"description_index\": 1,\n      \"matching_ids\": [\n        5\n      ],\n      \"start_frame\": 1,\n      \"end_frame\": 10,\n      \"confidence\": 1.0\n    },\n    {\n      \"description_index\": 2,\n      \"matching_ids\": [\n        6\n      ],\n      \"start_frame\": 1,\n      \"end_frame\": 4,\n      \"confidence\": 1.0\n    },\n    {\n      \"description_index\": 3,\n      \"matching_ids\": [],\n      \"start_frame\": 1,\n      \"end_frame\": 12,\n      \"confidence\": 1.0\n    }\n  ]\n}","raw_answers":[{"description_index":1,"matching_ids":[5],"start_frame":1,"end_frame":10,"confidence":1.0},{"description_index":2,"matching_ids":[6],"start_frame":1,"end_frame":4,"confidence":1.0},{"description_index":3,"matching_ids":[],"start_frame":1,"end_frame":12,"confidence":1.0}],"answers":[{"expression":"the stone balcony with the ornate green metal railing","matching_presentation_ids":[5],"start_choice_frame":1,"end_choice_frame":10,"confidence":1.0,"self_reported_confidence":1.0},{"expression":"the green wooden double door on the stone building","matching_presentation_ids":[6],"start_choice_frame":1,"end_choice_frame":4,"confidence":1.0,"self_reported_confidence":1.0},{"expression":"the white car parked down the cobblestone street","matching_presentation_ids":[],"start_choice_frame":1,"end_choice_frame":12,"confidence":1.0,"self_reported_confidence":1.0}]}},"verify":{"inputs":{"expressions":3,"tracklets":7,"selections":3,"gates":"interval, count, unknown ids, confidence, then geometry (motion)"},"outputs":{"kept":[{"expression":"the stone balcony with the ornate green metal railing","kind":"appearance","is_negative":false,"tracks":[{"timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36],"boxes":[[0.6104167103767395,0.1268518567085266,0.9994792342185974,0.6101851761341095],[0.5973958373069763,0.12870369851589203,0.999479204416275,0.6009259074926376],[0.5770833492279053,0.1518518477678299,0.999479204416275,0.588888868689537],[0.557812511920929,0.1814814805984497,0.999479204416275,0.577777773141861],[0.5239583849906921,0.23055554926395416,0.9994792342185974,0.5805555433034897],[0.48229169845581055,0.21111111342906952,0.9994792342185974,0.5898147970438004],[0.4281250238418579,0.14814814925193787,0.9130208790302277,0.5666666626930237],[0.37187501788139343,0.07222221791744232,0.8312500417232513,0.5259259194135666],[0.31041666865348816,0.03055555559694767,0.7598958611488342,0.4648148212581873],[0.2510416805744171,0.029629629105329514,0.7067708671092987,0.4027777723968029],[0.19427084922790527,0.029629629105329514,0.6671875417232513,0.3481481485068798],[0.15677084028720856,0.029629629105329514,0.6390625387430191,0.2888888828456402],[0.1848958432674408,0.029629629105329514,0.6145833730697632,0.24537036940455437],[0.23906251788139343,0.03055555559694767,0.5843750238418579,0.23518518544733524],[0.23333334922790527,0.03055555559694767,0.5625000298023224,0.2537036929279566],[0.2213541716337204,0.03055555559694767,0.5500000268220901,0.2694444451481104],[0.20520834624767303,0.03055555559694767,0.5510417073965073,0.281481483951211],[0.10000000894069672,0.03055555559694767,0.5843750387430191,0.3259259257465601],[0.19583334028720856,0.031481482088565826,0.6557291895151138,0.5111111178994179],[0.3072916865348816,0.3027777671813965,0.7739583849906921,0.7509258985519409],[0.3604166805744171,0.5064814686775208,0.869791716337204,0.9296296238899231],[0.3843750059604645,0.5370370149612427,0.9770833551883698,0.970370352268219],[0.38697919249534607,0.574999988079071,0.999479204416275,0.9712962806224823],[0.3739583492279053,0.5962963104248047,0.9994792342185974,0.972222238779068],[0.37552085518836975,0.5981481671333313,0.999479204416275,0.972222238779068],[0.3697916865348816,0.5916666388511658,0.9994792342185974,0.9712962508201599],[0.36250001192092896,0.5277777910232544,0.9994792342185974,0.9712963104248047],[0.3125000298023224,0.5129629373550415,0.999479204416275,0.9712962508201599],[0.20364584028720856,0.4888888895511627,0.8572917133569717,0.9722222089767456],[0.11041667312383652,0.4037036895751953,0.7546875104308128,0.9712962508201599],[0.05260416865348816,0.32499998807907104,0.6958333551883698,0.9712962508201599],[0.04010416939854622,0.27129629254341125,0.6781250424683094,0.9703703820705414],[0.08906250447034836,0.2796296179294586,0.7046875283122063,0.9694444239139557],[0.18854168057441711,0.38333332538604736,0.8161458671092987,0.970370352268219],[0.3125000298023224,0.4120370149612427,0.9651042520999908,0.970370352268219],[0.41979169845581055,0.29629629850387573,0.9994792342185974,0.8185185194015503],[0.6250000596046448,0.029629629105329514,0.9994792342185974,0.46481479331851006]]}]},{"expression":"the green wooden double door on the stone building","kind":"appearance","is_negative":false,"tracks":[{"timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12],"boxes":[[0.7359375357627869,0.8898147940635681,0.9630208760499954,0.9712962731719017],[0.7088541984558105,0.8768518567085266,0.9281250387430191,0.9712963029742241],[0.6677083969116211,0.8638888597488403,0.8713542371988297,0.9712962657213211],[0.6296875476837158,0.854629635810852,0.8203125596046448,0.9722222238779068],[0.5796875357627869,0.845370352268219,0.7562500387430191,0.9712962806224823],[0.5255208611488342,0.8185185194015503,0.6906250417232513,0.9712962955236435],[0.452604204416275,0.75,0.994791716337204,0.9712962955236435],[0.38229167461395264,0.690740704536438,0.542187511920929,0.9712962508201599],[0.06041666865348816,0.6111111044883728,0.4791666865348816,0.9722222089767456],[0.25468751788139343,0.5249999761581421,0.42760419845581055,0.9722221791744232],[0.20468750596046448,0.4592592418193817,0.6578125357627869,0.9990740716457367],[0.1744791716337204,0.39351850748062134,0.6442708522081375,0.9722222089767456],[0.13229167461395264,0.34166666865348816,0.6031250357627869,0.9712962806224823]]}]},{"expression":"the white car parked down the cobblestone street","kind":"spatial","is_negative":true,"tracks":[]}],"rejected":{},"rejection_counts":{},"row":{"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_011_scene_019.mp4","scene_key":"--cif7Nv6BI_chunk_011_scene_019","category":"test_vehicles","height":1080,"width":1920,"fps":50.0,"num_frames":577,"duration":11.540000000000001,"rmot_expressions":[{"expression":"the stone balcony with the ornate green metal railing","kind":"appearance","is_negative":false,"tracks":[{"timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36],"boxes":[[0.6104167103767395,0.1268518567085266,0.9994792342185974,0.6101851761341095],[0.5973958373069763,0.12870369851589203,0.999479204416275,0.6009259074926376],[0.5770833492279053,0.1518518477678299,0.999479204416275,0.588888868689537],[0.557812511920929,0.1814814805984497,0.999479204416275,0.577777773141861],[0.5239583849906921,0.23055554926395416,0.9994792342185974,0.5805555433034897],[0.48229169845581055,0.21111111342906952,0.9994792342185974,0.5898147970438004],[0.4281250238418579,0.14814814925193787,0.9130208790302277,0.5666666626930237],[0.37187501788139343,0.07222221791744232,0.8312500417232513,0.5259259194135666],[0.31041666865348816,0.03055555559694767,0.7598958611488342,0.4648148212581873],[0.2510416805744171,0.029629629105329514,0.7067708671092987,0.4027777723968029],[0.19427084922790527,0.029629629105329514,0.6671875417232513,0.3481481485068798],[0.15677084028720856,0.029629629105329514,0.6390625387430191,0.2888888828456402],[0.1848958432674408,0.029629629105329514,0.6145833730697632,0.24537036940455437],[0.23906251788139343,0.03055555559694767,0.5843750238418579,0.23518518544733524],[0.23333334922790527,0.03055555559694767,0.5625000298023224,0.2537036929279566],[0.2213541716337204,0.03055555559694767,0.5500000268220901,0.2694444451481104],[0.20520834624767303,0.03055555559694767,0.5510417073965073,0.281481483951211],[0.10000000894069672,0.03055555559694767,0.5843750387430191,0.3259259257465601],[0.19583334028720856,0.031481482088565826,0.6557291895151138,0.5111111178994179],[0.3072916865348816,0.3027777671813965,0.7739583849906921,0.7509258985519409],[0.3604166805744171,0.5064814686775208,0.869791716337204,0.9296296238899231],[0.3843750059604645,0.5370370149612427,0.9770833551883698,0.970370352268219],[0.38697919249534607,0.574999988079071,0.999479204416275,0.9712962806224823],[0.3739583492279053,0.5962963104248047,0.9994792342185974,0.972222238779068],[0.37552085518836975,0.5981481671333313,0.999479204416275,0.972222238779068],[0.3697916865348816,0.5916666388511658,0.9994792342185974,0.9712962508201599],[0.36250001192092896,0.5277777910232544,0.9994792342185974,0.9712963104248047],[0.3125000298023224,0.5129629373550415,0.999479204416275,0.9712962508201599],[0.20364584028720856,0.4888888895511627,0.8572917133569717,0.9722222089767456],[0.11041667312383652,0.4037036895751953,0.7546875104308128,0.9712962508201599],[0.05260416865348816,0.32499998807907104,0.6958333551883698,0.9712962508201599],[0.04010416939854622,0.27129629254341125,0.6781250424683094,0.9703703820705414],[0.08906250447034836,0.2796296179294586,0.7046875283122063,0.9694444239139557],[0.18854168057441711,0.38333332538604736,0.8161458671092987,0.970370352268219],[0.3125000298023224,0.4120370149612427,0.9651042520999908,0.970370352268219],[0.41979169845581055,0.29629629850387573,0.9994792342185974,0.8185185194015503],[0.6250000596046448,0.029629629105329514,0.9994792342185974,0.46481479331851006]]}]},{"expression":"the green wooden double door on the stone building","kind":"appearance","is_negative":false,"tracks":[{"timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12],"boxes":[[0.7359375357627869,0.8898147940635681,0.9630208760499954,0.9712962731719017],[0.7088541984558105,0.8768518567085266,0.9281250387430191,0.9712963029742241],[0.6677083969116211,0.8638888597488403,0.8713542371988297,0.9712962657213211],[0.6296875476837158,0.854629635810852,0.8203125596046448,0.9722222238779068],[0.5796875357627869,0.845370352268219,0.7562500387430191,0.9712962806224823],[0.5255208611488342,0.8185185194015503,0.6906250417232513,0.9712962955236435],[0.452604204416275,0.75,0.994791716337204,0.9712962955236435],[0.38229167461395264,0.690740704536438,0.542187511920929,0.9712962508201599],[0.06041666865348816,0.6111111044883728,0.4791666865348816,0.9722222089767456],[0.25468751788139343,0.5249999761581421,0.42760419845581055,0.9722221791744232],[0.20468750596046448,0.4592592418193817,0.6578125357627869,0.9990740716457367],[0.1744791716337204,0.39351850748062134,0.6442708522081375,0.9722222089767456],[0.13229167461395264,0.34166666865348816,0.6031250357627869,0.9712962806224823]]}]},{"expression":"the white car parked down the cobblestone street","kind":"spatial","is_negative":true,"tracks":[]}],"rmot_expression_count":3,"rmot_negative_count":1,"rmot_motion_count":0}}}}},{"index":15,"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_011_scene_001.mp4","status":"complete","steps":{"decode":{"inputs":{"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_011_scene_001.mp4","sample_fps":4.0},"outputs":{"width":1920,"height":1080,"source_fps":50.0,"duration_seconds":29.560000000000002,"frames_kept":114,"blank_frames_dropped":0,"timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88,10.14,10.4,10.66,10.92,11.18,11.44,11.700000000000001,11.96,12.22,12.48,12.74,13.0,13.26,13.52,13.780000000000001,14.040000000000001,14.3,14.56,14.82,15.08,15.34,15.6,15.860000000000001,16.12,16.38,16.64,16.9,17.16,17.42,17.68,17.94,18.2,18.46,18.72,18.98,19.240000000000002,19.5,19.76,20.02,20.28,20.54,20.8,21.06,21.32,21.580000000000002,21.84,22.1,22.36,22.62,22.88,23.14,23.400000000000002,23.66,23.92,24.18,24.44,24.7,24.96,25.22,25.48,25.740000000000002,26.0,26.26,26.52,26.78,27.04,27.3,27.560000000000002,27.82,28.080000000000002,28.34,28.6,28.86,29.12,29.38],"error":""}},"propose":{"inputs":{"model":"gemini-3.5-flash-lite","temperature":0.4,"video_bytes":48078773,"system_prompt":"You are annotating video clips to build training data for referring multi-object tracking: given a video and a natural-language description, a model must output the boxes of every object matching that description, over time.\n\nYour job is to write the descriptions. You do NOT output coordinates, boxes, pixel positions, or any numbers about where things are. A separate detector finds the objects.\n\nFor every clip you produce a list of items. Each item has two texts that must NOT be the same:\n\nEXPRESSION — what a person would say to single out the object or objects. It may describe motion, temporal order, interaction with other objects, and spatial relations. It should read like something a person would actually say.\n  Good: \"the black car that's turning left\"\n  Good: \"the cyclist who overtakes the bus\"\n  Good: \"the two people who walk out of frame on the right\"\n\nNOUN_PHRASE — the same object as a plain, timeless visual category with its visible attributes, and NOTHING else. The detector is prompted with this text and looks at ONE frame at a time, so anything about time, order, or action is useless to it and makes it fail.\n  Allowed: category nouns, colours, materials, clothing, size, type\n  Forbidden: verbs and participles of action (\"turning\", \"walks\", \"is moving\"), temporal words (\"then\", \"starts\", \"after\", \"first\", \"last\"), relative clauses (\"that ...\", \"which ...\", \"who ...\"), ordinals and rank words (\"first\", \"leftmost\", \"nearest\"), and anything over six words.\n  \"the black car that's turning left\" -> \"black car\"\n  \"the cyclist who overtakes the bus\" -> \"cyclist\"\n  \"the two people who walk out of frame on the right\" -> \"person\"\n\nKIND — what makes the expression hard: \"motion\", \"appearance\", \"spatial\", or \"interaction\". Aim for about 35% \"motion\" items: motion is the part a single-frame model cannot already solve, so it is the most valuable, but a list of only motion items is not usable training data either.\n\nEXPECTED_COUNT — how many distinct objects in the clip match the EXPRESSION. Count carefully; this is checked against the detector's answer, and a wrong count discards the item.\n\nIS_NEGATIVE — at most ONE item may be a negative, and only when you also write at least one non-negative item. A negative is an expression that recombines attributes that ARE present in the clip into a whole that is NOT. If the only black car is moving, \"the parked black car\" is a good negative. Its NOUN_PHRASE must still be the category that IS present (\"black car\"), because that is what makes the absence checkable rather than merely unverified. A negative's EXPECTED_COUNT is 0.\n\nRules that discard an item, so follow them exactly:\n- Never write coordinates, box values, pixel positions, or percentages of the frame.\n- Never let a NOUN_PHRASE contain a verb of action, a temporal word, a relative clause, or an ordinal.\n- Never write an EXPRESSION whose object is not visible in the clip, unless it is marked IS_NEGATIVE.\n- Prefer objects that are clearly visible and identifiable over background clutter.\n- Do not describe the camera or the shot; describe objects in the scene.\n","user_prompt":"Watch this clip and write between 1 and 3 items following the rules. Cover different objects rather than restating one object several ways, and include any negative in the same list.\n"},"outputs":{"raw_response":"{\n  \"items\": [\n    {\n      \"expression\": \"the person wearing jeans who walks towards the stone structure\",\n      \"noun_phrase\": \"person\",\n      \"kind\": \"motion\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the stone coffin carved into the rock\",\n      \"noun_phrase\": \"stone coffin\",\n      \"kind\": \"appearance\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    }\n  ]\n}","raw_items":[{"expression":"the person wearing jeans who walks towards the stone structure","noun_phrase":"person","kind":"motion","expected_count":1,"is_negative":false},{"expression":"the stone coffin carved into the rock","noun_phrase":"stone coffin","kind":"appearance","expected_count":1,"is_negative":false}],"accepted":[{"expression":"the person wearing jeans who walks towards the stone structure","noun_phrase":"person","kind":"motion","expected_count":1,"is_negative":false},{"expression":"the stone coffin carved into the rock","noun_phrase":"stone coffin","kind":"appearance","expected_count":1,"is_negative":false}],"gated":[]}},"track":{"inputs":{"tracker":"Sam31MultiplexTracker","model_id":"/mnt/ceph3/ec/jacky/sam3.1","noun_phrases":["person","stone coffin"],"expressions_in":2,"frames":114,"sample_fps":4.0,"frame_size":"1920x1080","timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88,10.14,10.4,10.66,10.92,11.18,11.44,11.700000000000001,11.96,12.22,12.48,12.74,13.0,13.26,13.52,13.780000000000001,14.040000000000001,14.3,14.56,14.82,15.08,15.34,15.6,15.860000000000001,16.12,16.38,16.64,16.9,17.16,17.42,17.68,17.94,18.2,18.46,18.72,18.98,19.240000000000002,19.5,19.76,20.02,20.28,20.54,20.8,21.06,21.32,21.580000000000002,21.84,22.1,22.36,22.62,22.88,23.14,23.400000000000002,23.66,23.92,24.18,24.44,24.7,24.96,25.22,25.48,25.740000000000002,26.0,26.26,26.52,26.78,27.04,27.3,27.560000000000002,27.82,28.080000000000002,28.34,28.6,28.86,29.12,29.38],"session":"one session over 114 JPEG frames; then per phrase: reset_session, add_prompt(text) on frame 0, propagate_in_video","passes":2},"outputs":{"raw_tracks":[{"noun_phrase":"person","object_id":0,"observations":[{"timestamp":0.0,"box":[0.5911458730697632,0.46481481194496155,0.6312500424683094,0.7166666686534882],"occluded":false},{"timestamp":0.26,"box":[0.581250011920929,0.47592592239379883,0.6265625134110451,0.7407407462596893],"occluded":false},{"timestamp":0.52,"box":[0.5807291865348816,0.5361111164093018,0.6244791895151138,0.8287037014961243],"occluded":false},{"timestamp":0.78,"box":[0.5937500596046448,0.5175926089286804,0.637500062584877,0.8166666924953461],"occluded":false},{"timestamp":1.04,"box":[0.5619791746139526,0.5120370388031006,0.6171875111758709,0.8314814865589142],"occluded":false},{"timestamp":1.3,"box":[0.5291666984558105,0.4981481432914734,0.5828125327825546,0.8296296298503876],"occluded":false},{"timestamp":1.56,"box":[0.5453125238418579,0.4842592477798462,0.6010416932404041,0.8388888835906982],"occluded":false},{"timestamp":1.82,"box":[0.5380208492279053,0.49351850152015686,0.5984375178813934,0.8657407164573669],"occluded":false},{"timestamp":2.08,"box":[0.48489585518836975,0.47777777910232544,0.5463541932404041,0.864814817905426],"occluded":false},{"timestamp":2.34,"box":[0.47083336114883423,0.4694444239139557,0.5375000312924385,0.8907407224178314],"occluded":false},{"timestamp":2.6,"box":[0.4635416865348816,0.4749999940395355,0.5395833551883698,0.9342592358589172],"occluded":false},{"timestamp":2.86,"box":[0.4593750238418579,0.41111111640930176,0.544270858168602,0.9000000059604645],"occluded":false},{"timestamp":3.12,"box":[0.4322916865348816,0.43611109256744385,0.5302083566784859,0.9712962508201599],"occluded":false},{"timestamp":3.38,"box":[0.4010416865348816,0.41111111640930176,0.49947918951511383,0.970370352268219],"occluded":false},{"timestamp":3.64,"box":[0.40781253576278687,0.42314815521240234,0.4958333745598793,0.9722222089767456],"occluded":false},{"timestamp":3.9,"box":[0.4411458671092987,0.35555553436279297,0.5338542088866234,0.9731481075286865],"occluded":false},{"timestamp":4.16,"box":[0.41718751192092896,0.35555553436279297,0.5296875163912773,0.9712962508201599],"occluded":false},{"timestamp":4.42,"box":[0.3880208432674408,0.2953703701496124,0.526562511920929,0.972222238779068],"occluded":false},{"timestamp":4.68,"box":[0.3421875238418579,0.2212962955236435,0.5369791984558105,0.9712962955236435],"occluded":false},{"timestamp":4.94,"box":[0.26354166865348816,0.04444444179534912,0.5432291924953461,0.9712962508201599],"occluded":false},{"timestamp":5.2,"box":[0.10104167461395264,0.031481482088565826,0.4921875298023224,0.9722221866250038],"occluded":false},{"timestamp":5.46,"box":[0.0,0.028703702613711357,0.47447919845581055,0.9759259354323149],"occluded":false},{"timestamp":5.72,"box":[0.0,0.028703702613711357,0.4401041865348816,0.9731481205672026],"occluded":false},{"timestamp":5.98,"box":[0.0,0.0,0.3333333432674408,0.9722222089767456],"occluded":false},{"timestamp":6.24,"box":[0.0,0.0,0.29895836114883423,0.9722222089767456],"occluded":false},{"timestamp":6.5,"box":[0.0,0.03055555559694767,0.2838541865348816,0.9722222182899714],"occluded":false},{"timestamp":6.76,"box":[0.0,0.0416666641831398,0.19114583730697632,0.9722221866250038],"occluded":false},{"timestamp":13.0,"box":[0.0,0.4203703701496124,0.26354166865348816,0.9907407462596893],"occluded":false}],"scores":[0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919,0.8987342119216919]},{"noun_phrase":"person","object_id":1,"observations":[{"timestamp":13.26,"box":[0.0,0.11296296119689941,0.2682291865348816,0.9712962508201599],"occluded":false},{"timestamp":13.52,"box":[0.0,0.028703702613711357,0.2640625238418579,0.8990740310400724],"occluded":false},{"timestamp":13.780000000000001,"box":[0.0,0.026851851493120193,0.24687501788139343,0.7861110754311085],"occluded":false},{"timestamp":14.040000000000001,"box":[0.0,0.026851851493120193,0.22812500596046448,0.7490740604698658],"occluded":false},{"timestamp":14.3,"box":[0.0,0.026851851493120193,0.2265625149011612,0.7916666455566883],"occluded":false},{"timestamp":14.56,"box":[0.0,0.026851851493120193,0.22604167461395264,0.7861110754311085],"occluded":false},{"timestamp":14.82,"box":[0.0,0.026851851493120193,0.23750001192092896,0.7907407470047474],"occluded":false},{"timestamp":15.08,"box":[0.0,0.025925925001502037,0.24375000596046448,0.79629628919065],"occluded":false},{"timestamp":15.34,"box":[0.0,0.02777777798473835,0.24687501788139343,0.8268518317490816],"occluded":false},{"timestamp":15.6,"box":[0.0,0.028703702613711357,0.2161458432674408,0.9990740548819304],"occluded":false},{"timestamp":15.860000000000001,"box":[0.0,0.635185182094574,0.06927083432674408,0.9203703701496124],"occluded":false},{"timestamp":17.42,"box":[0.0,0.8685185313224792,0.013541667722165585,0.9722222313284874],"occluded":false},{"timestamp":17.68,"box":[0.0,0.5157407522201538,0.11666667461395264,0.972222238779068],"occluded":false},{"timestamp":17.94,"box":[0.0,0.17037037014961243,0.18541668355464935,0.9990740716457367],"occluded":false},{"timestamp":18.2,"box":[0.0,0.04722222313284874,0.20208334922790527,0.9305555485188961],"occluded":false},{"timestamp":18.46,"box":[0.0,0.03055555559694767,0.2109375149011612,0.9212963078171015],"occluded":false},{"timestamp":18.72,"box":[0.0,0.03055555559694767,0.20781251788139343,0.9212963078171015],"occluded":false},{"timestamp":18.98,"box":[0.0,0.032407406717538834,0.20625001192092896,0.9407407082617283],"occluded":false},{"timestamp":19.240000000000002,"box":[0.0,0.031481482088565826,0.20729167759418488,0.9361110702157021],"occluded":false},{"timestamp":19.5,"box":[0.0,0.029629629105329514,0.20625001192092896,0.9342592172324657],"occluded":false},{"timestamp":19.76,"box":[0.0,0.029629629105329514,0.20781251788139343,0.947222214192152],"occluded":false},{"timestamp":20.02,"box":[0.0,0.029629629105329514,0.20885418355464935,0.9601851515471935],"occluded":false},{"timestamp":20.28,"box":[0.0,0.029629629105329514,0.2057291716337204,0.9638888649642467],"occluded":false},{"timestamp":20.54,"box":[0.0,0.028703702613711357,0.19843751192092896,0.9333332907408476],"occluded":false},{"timestamp":20.8,"box":[0.0,0.03055555559694767,0.24114584922790527,0.8898148033767939],"occluded":false},{"timestamp":21.06,"box":[0.0,0.10648147761821747,0.21041667461395264,0.859259232878685],"occluded":false},{"timestamp":21.32,"box":[0.0,0.520370364189148,0.10052084177732468,0.9314814805984497],"occluded":false}],"scores":[0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562]},{"noun_phrase":"person","object_id":2,"observations":[{"timestamp":25.48,"box":[0.3291666805744171,0.7388888597488403,0.4828125238418579,0.9712962657213211],"occluded":false},{"timestamp":25.740000000000002,"box":[0.10468750447034836,0.6620370149612427,0.4286458417773247,0.9712962806224823],"occluded":false},{"timestamp":26.0,"box":[0.23750001192092896,0.7509258985519409,0.48489585518836975,0.9712962657213211],"occluded":false},{"timestamp":26.26,"box":[0.2562500238418579,0.6620370149612427,0.5489583611488342,0.9722222089767456],"occluded":false},{"timestamp":26.52,"box":[0.2291666716337204,0.845370352268219,0.4270833432674408,0.9712962806224823],"occluded":false}],"scores":[0.8012422323226929,0.8012422323226929,0.8012422323226929,0.8012422323226929,0.8012422323226929]}],"tracklets":[{"noun_phrase":"person","object_id":0,"presentation_id":1,"observations":[{"timestamp":0.0,"box":[0.5911458730697632,0.46481481194496155,0.6312500424683094,0.7166666686534882],"occluded":false},{"timestamp":0.26,"box":[0.581250011920929,0.47592592239379883,0.6265625134110451,0.7407407462596893],"occluded":false},{"timestamp":0.52,"box":[0.5807291865348816,0.5361111164093018,0.6244791895151138,0.8287037014961243],"occluded":false},{"timestamp":0.78,"box":[0.5937500596046448,0.5175926089286804,0.637500062584877,0.8166666924953461],"occluded":false},{"timestamp":1.04,"box":[0.5619791746139526,0.5120370388031006,0.6171875111758709,0.8314814865589142],"occluded":false},{"timestamp":1.3,"box":[0.5291666984558105,0.4981481432914734,0.5828125327825546,0.8296296298503876],"occluded":false},{"timestamp":1.56,"box":[0.5453125238418579,0.4842592477798462,0.6010416932404041,0.8388888835906982],"occluded":false},{"timestamp":1.82,"box":[0.5380208492279053,0.49351850152015686,0.5984375178813934,0.8657407164573669],"occluded":false},{"timestamp":2.08,"box":[0.48489585518836975,0.47777777910232544,0.5463541932404041,0.864814817905426],"occluded":false},{"timestamp":2.34,"box":[0.47083336114883423,0.4694444239139557,0.5375000312924385,0.8907407224178314],"occluded":false},{"timestamp":2.6,"box":[0.4635416865348816,0.4749999940395355,0.5395833551883698,0.9342592358589172],"occluded":false},{"timestamp":2.86,"box":[0.4593750238418579,0.41111111640930176,0.544270858168602,0.9000000059604645],"occluded":false},{"timestamp":3.12,"box":[0.4322916865348816,0.43611109256744385,0.5302083566784859,0.9712962508201599],"occluded":false},{"timestamp":3.38,"box":[0.4010416865348816,0.41111111640930176,0.49947918951511383,0.970370352268219],"occluded":false},{"timestamp":3.64,"box":[0.40781253576278687,0.42314815521240234,0.4958333745598793,0.9722222089767456],"occluded":false},{"timestamp":3.9,"box":[0.4411458671092987,0.35555553436279297,0.5338542088866234,0.9731481075286865],"occluded":false},{"timestamp":4.16,"box":[0.41718751192092896,0.35555553436279297,0.5296875163912773,0.9712962508201599],"occluded":false},{"timestamp":4.42,"box":[0.3880208432674408,0.2953703701496124,0.526562511920929,0.972222238779068],"occluded":false},{"timestamp":4.68,"box":[0.3421875238418579,0.2212962955236435,0.5369791984558105,0.9712962955236435],"occluded":false},{"timestamp":4.94,"box":[0.26354166865348816,0.04444444179534912,0.5432291924953461,0.9712962508201599],"occluded":false},{"timestamp":5.2,"box":[0.10104167461395264,0.031481482088565826,0.4921875298023224,0.9722221866250038],"occluded":false},{"timestamp":5.46,"box":[0.0,0.028703702613711357,0.47447919845581055,0.9759259354323149],"occluded":false},{"timestamp":5.72,"box":[0.0,0.028703702613711357,0.4401041865348816,0.9731481205672026],"occluded":false},{"timestamp":5.98,"box":[0.0,0.0,0.3333333432674408,0.9722222089767456],"occluded":false},{"timestamp":6.24,"box":[0.0,0.0,0.29895836114883423,0.9722222089767456],"occluded":false},{"timestamp":6.5,"box":[0.0,0.03055555559694767,0.2838541865348816,0.9722222182899714],"occluded":false},{"timestamp":6.76,"box":[0.0,0.0416666641831398,0.19114583730697632,0.9722221866250038],"occluded":false},{"timestamp":13.0,"box":[0.0,0.4203703701496124,0.26354166865348816,0.9907407462596893],"occluded":false}],"mean_score":0.8987342119216919},{"noun_phrase":"person","object_id":2,"presentation_id":2,"observations":[{"timestamp":25.48,"box":[0.3291666805744171,0.7388888597488403,0.4828125238418579,0.9712962657213211],"occluded":false},{"timestamp":25.740000000000002,"box":[0.10468750447034836,0.6620370149612427,0.4286458417773247,0.9712962806224823],"occluded":false},{"timestamp":26.0,"box":[0.23750001192092896,0.7509258985519409,0.48489585518836975,0.9712962657213211],"occluded":false},{"timestamp":26.26,"box":[0.2562500238418579,0.6620370149612427,0.5489583611488342,0.9722222089767456],"occluded":false},{"timestamp":26.52,"box":[0.2291666716337204,0.845370352268219,0.4270833432674408,0.9712962806224823],"occluded":false}],"mean_score":0.8012422323226929},{"noun_phrase":"person","object_id":1,"presentation_id":3,"observations":[{"timestamp":13.26,"box":[0.0,0.11296296119689941,0.2682291865348816,0.9712962508201599],"occluded":false},{"timestamp":13.52,"box":[0.0,0.028703702613711357,0.2640625238418579,0.8990740310400724],"occluded":false},{"timestamp":13.780000000000001,"box":[0.0,0.026851851493120193,0.24687501788139343,0.7861110754311085],"occluded":false},{"timestamp":14.040000000000001,"box":[0.0,0.026851851493120193,0.22812500596046448,0.7490740604698658],"occluded":false},{"timestamp":14.3,"box":[0.0,0.026851851493120193,0.2265625149011612,0.7916666455566883],"occluded":false},{"timestamp":14.56,"box":[0.0,0.026851851493120193,0.22604167461395264,0.7861110754311085],"occluded":false},{"timestamp":14.82,"box":[0.0,0.026851851493120193,0.23750001192092896,0.7907407470047474],"occluded":false},{"timestamp":15.08,"box":[0.0,0.025925925001502037,0.24375000596046448,0.79629628919065],"occluded":false},{"timestamp":15.34,"box":[0.0,0.02777777798473835,0.24687501788139343,0.8268518317490816],"occluded":false},{"timestamp":15.6,"box":[0.0,0.028703702613711357,0.2161458432674408,0.9990740548819304],"occluded":false},{"timestamp":15.860000000000001,"box":[0.0,0.635185182094574,0.06927083432674408,0.9203703701496124],"occluded":false},{"timestamp":17.42,"box":[0.0,0.8685185313224792,0.013541667722165585,0.9722222313284874],"occluded":false},{"timestamp":17.68,"box":[0.0,0.5157407522201538,0.11666667461395264,0.972222238779068],"occluded":false},{"timestamp":17.94,"box":[0.0,0.17037037014961243,0.18541668355464935,0.9990740716457367],"occluded":false},{"timestamp":18.2,"box":[0.0,0.04722222313284874,0.20208334922790527,0.9305555485188961],"occluded":false},{"timestamp":18.46,"box":[0.0,0.03055555559694767,0.2109375149011612,0.9212963078171015],"occluded":false},{"timestamp":18.72,"box":[0.0,0.03055555559694767,0.20781251788139343,0.9212963078171015],"occluded":false},{"timestamp":18.98,"box":[0.0,0.032407406717538834,0.20625001192092896,0.9407407082617283],"occluded":false},{"timestamp":19.240000000000002,"box":[0.0,0.031481482088565826,0.20729167759418488,0.9361110702157021],"occluded":false},{"timestamp":19.5,"box":[0.0,0.029629629105329514,0.20625001192092896,0.9342592172324657],"occluded":false},{"timestamp":19.76,"box":[0.0,0.029629629105329514,0.20781251788139343,0.947222214192152],"occluded":false},{"timestamp":20.02,"box":[0.0,0.029629629105329514,0.20885418355464935,0.9601851515471935],"occluded":false},{"timestamp":20.28,"box":[0.0,0.029629629105329514,0.2057291716337204,0.9638888649642467],"occluded":false},{"timestamp":20.54,"box":[0.0,0.028703702613711357,0.19843751192092896,0.9333332907408476],"occluded":false},{"timestamp":20.8,"box":[0.0,0.03055555559694767,0.24114584922790527,0.8898148033767939],"occluded":false},{"timestamp":21.06,"box":[0.0,0.10648147761821747,0.21041667461395264,0.859259232878685],"occluded":false},{"timestamp":21.32,"box":[0.0,0.520370364189148,0.10052084177732468,0.9314814805984497],"occluded":false}],"mean_score":0.9609756469726562}],"dropped":{},"fates":[{"noun_phrase":"person","object_id":0,"frames":28,"mean_score":0.8987342119216919,"mean_area":0.1242369079622324,"outcome":"kept","presentation_id":1,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"person","object_id":1,"frames":27,"mean_score":0.9609756469726562,"mean_area":0.16437345234682924,"outcome":"kept","presentation_id":3,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"person","object_id":2,"frames":5,"mean_score":0.8012422323226929,"mean_area":0.06122617903988763,"outcome":"kept","presentation_id":2,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null}],"gates":{"min_track_frames":3,"min_mean_score":0.5,"merge_iou":0.8,"max_tracklets":12}}},"render":{"inputs":{"tracklets":3,"as_video":true,"fps":1.0,"max_frames":30},"outputs":{"media":"/home/zzk/data/video_bbox/vlm_gemini3.5_lite/renders/--cif7Nv6BI_chunk_011_scene_001_choices.mp4","media_bytes":12575681,"frames":[{"index":1,"timestamp":0.0,"file":"--cif7Nv6BI_chunk_011_scene_001_choices.mp4"},{"index":2,"timestamp":1.04,"file":"--cif7Nv6BI_chunk_011_scene_001_choices.mp4"},{"index":3,"timestamp":2.08,"file":"--cif7Nv6BI_chunk_011_scene_001_choices.mp4"},{"index":4,"timestamp":3.12,"file":"--cif7Nv6BI_chunk_011_scene_001_choices.mp4"},{"index":5,"timestamp":4.16,"file":"--cif7Nv6BI_chunk_011_scene_001_choices.mp4"},{"index":6,"timestamp":4.94,"file":"--cif7Nv6BI_chunk_011_scene_001_choices.mp4"},{"index":7,"timestamp":5.98,"file":"--cif7Nv6BI_chunk_011_scene_001_choices.mp4"},{"index":8,"timestamp":7.0200000000000005,"file":"--cif7Nv6BI_chunk_011_scene_001_choices.mp4"},{"index":9,"timestamp":8.06,"file":"--cif7Nv6BI_chunk_011_scene_001_choices.mp4"},{"index":10,"timestamp":9.1,"file":"--cif7Nv6BI_chunk_011_scene_001_choices.mp4"},{"index":11,"timestamp":10.14,"file":"--cif7Nv6BI_chunk_011_scene_001_choices.mp4"},{"index":12,"timestamp":11.18,"file":"--cif7Nv6BI_chunk_011_scene_001_choices.mp4"},{"index":13,"timestamp":12.22,"file":"--cif7Nv6BI_chunk_011_scene_001_choices.mp4"},{"index":14,"timestamp":13.26,"file":"--cif7Nv6BI_chunk_011_scene_001_choices.mp4"},{"index":15,"timestamp":14.3,"file":"--cif7Nv6BI_chunk_011_scene_001_choices.mp4"},{"index":16,"timestamp":15.08,"file":"--cif7Nv6BI_chunk_011_scene_001_choices.mp4"},{"index":17,"timestamp":16.12,"file":"--cif7Nv6BI_chunk_011_scene_001_choices.mp4"},{"index":18,"timestamp":17.16,"file":"--cif7Nv6BI_chunk_011_scene_001_choices.mp4"},{"index":19,"timestamp":18.2,"file":"--cif7Nv6BI_chunk_011_scene_001_choices.mp4"},{"index":20,"timestamp":19.240000000000002,"file":"--cif7Nv6BI_chunk_011_scene_001_choices.mp4"},{"index":21,"timestamp":20.28,"file":"--cif7Nv6BI_chunk_011_scene_001_choices.mp4"},{"index":22,"timestamp":21.32,"file":"--cif7Nv6BI_chunk_011_scene_001_choices.mp4"},{"index":23,"timestamp":22.36,"file":"--cif7Nv6BI_chunk_011_scene_001_choices.mp4"},{"index":24,"timestamp":23.400000000000002,"file":"--cif7Nv6BI_chunk_011_scene_001_choices.mp4"},{"index":25,"timestamp":24.44,"file":"--cif7Nv6BI_chunk_011_scene_001_choices.mp4"},{"index":26,"timestamp":25.22,"file":"--cif7Nv6BI_chunk_011_scene_001_choices.mp4"},{"index":27,"timestamp":26.26,"file":"--cif7Nv6BI_chunk_011_scene_001_choices.mp4"},{"index":28,"timestamp":27.3,"file":"--cif7Nv6BI_chunk_011_scene_001_choices.mp4"},{"index":29,"timestamp":28.34,"file":"--cif7Nv6BI_chunk_011_scene_001_choices.mp4"},{"index":30,"timestamp":29.38,"file":"--cif7Nv6BI_chunk_011_scene_001_choices.mp4"}]}},"select":{"inputs":{"model":"gemini-3.5-flash-lite","temperature":0.0,"as_video":true,"media_bytes":12575681,"media":["--cif7Nv6BI_chunk_011_scene_001_choices.mp4"],"choice_frames":30,"system_prompt":"You are labelling video data for referring multi-object tracking.\n\nYou will see a numbered sequence of frames from one clip, in time order. Each frame has its own number and timestamp printed in the top-left corner. Candidate objects have been detected for you and drawn on the frames as coloured boxes, each labelled with a number. The same object keeps the same number and the same colour on every frame. Below each frame is a strip of enlarged crops of those same numbered boxes, so you can identify a small or distant object.\n\nFor each description you are given, answer two questions:\n\n1. MATCHING_IDS — which of the numbered objects the description refers to. List every number that matches, or an EMPTY LIST if none of them do. An empty list is a correct and expected answer: some descriptions are deliberately about something that is not in this clip. Do not pick a number just to have picked one, and do not pick a number for an object that merely resembles the description.\n\n2. START_FRAME and END_FRAME — the frame numbers between which the described state actually holds. If the description is \"the car that's turning left\", the interval is the frames in which that car is turning, NOT every frame in which the car is visible. Use the frame numbers printed on the frames. START_FRAME must not be greater than END_FRAME. When MATCHING_IDS is empty, give the full range (the first and last frame numbers); it is ignored.\n\n3. CONFIDENCE — how sure you are of the MATCHING_IDS answer, from 0.0 to 1.0. Be honest: a low number costs the example nothing, and an overconfident wrong answer becomes a wrong label.\n\nRules:\n- Only use numbers that are actually drawn on the frames.\n- Never output coordinates, box values, or pixel positions.\n- Judge motion by comparing the object's position across frames, not by guessing from one frame.\n","user_prompt":"The video has 30 frames, numbered 1 to 30. Each frame is one moment of the clip and has its number burned into the top left corner.\n\nAnswer for each of the following descriptions, in the order given:\n1. the person wearing jeans who walks towards the stone structure\n2. the stone coffin carved into the rock","questions":["the person wearing jeans who walks towards the stone structure","the stone coffin carved into the rock"]},"outputs":{"raw_response":"{\n  \"answers\": [\n    {\n      \"description_index\": 1,\n      \"matching_ids\": [\n        1\n      ],\n      \"start_frame\": 1,\n      \"end_frame\": 6,\n      \"confidence\": 1.0\n    },\n    {\n      \"description_index\": 2,\n      \"matching_ids\": [],\n      \"start_frame\": 1,\n      \"end_frame\": 30,\n      \"confidence\": 1.0\n    }\n  ]\n}","raw_answers":[{"description_index":1,"matching_ids":[1],"start_frame":1,"end_frame":6,"confidence":1.0},{"description_index":2,"matching_ids":[],"start_frame":1,"end_frame":30,"confidence":1.0}],"answers":[{"expression":"the person wearing jeans who walks towards the stone structure","matching_presentation_ids":[1],"start_choice_frame":1,"end_choice_frame":6,"confidence":1.0,"self_reported_confidence":1.0},{"expression":"the stone coffin carved into the rock","matching_presentation_ids":[],"start_choice_frame":1,"end_choice_frame":30,"confidence":1.0,"self_reported_confidence":1.0}]}},"verify":{"inputs":{"expressions":2,"tracklets":3,"selections":2,"gates":"interval, count, unknown ids, confidence, then geometry (motion)"},"outputs":{"kept":[{"expression":"the person wearing jeans who walks towards the stone structure","kind":"motion","is_negative":false,"tracks":[{"timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94],"boxes":[[0.5911458730697632,0.46481481194496155,0.6312500424683094,0.7166666686534882],[0.581250011920929,0.47592592239379883,0.6265625134110451,0.7407407462596893],[0.5807291865348816,0.5361111164093018,0.6244791895151138,0.8287037014961243],[0.5937500596046448,0.5175926089286804,0.637500062584877,0.8166666924953461],[0.5619791746139526,0.5120370388031006,0.6171875111758709,0.8314814865589142],[0.5291666984558105,0.4981481432914734,0.5828125327825546,0.8296296298503876],[0.5453125238418579,0.4842592477798462,0.6010416932404041,0.8388888835906982],[0.5380208492279053,0.49351850152015686,0.5984375178813934,0.8657407164573669],[0.48489585518836975,0.47777777910232544,0.5463541932404041,0.864814817905426],[0.47083336114883423,0.4694444239139557,0.5375000312924385,0.8907407224178314],[0.4635416865348816,0.4749999940395355,0.5395833551883698,0.9342592358589172],[0.4593750238418579,0.41111111640930176,0.544270858168602,0.9000000059604645],[0.4322916865348816,0.43611109256744385,0.5302083566784859,0.9712962508201599],[0.4010416865348816,0.41111111640930176,0.49947918951511383,0.970370352268219],[0.40781253576278687,0.42314815521240234,0.4958333745598793,0.9722222089767456],[0.4411458671092987,0.35555553436279297,0.5338542088866234,0.9731481075286865],[0.41718751192092896,0.35555553436279297,0.5296875163912773,0.9712962508201599],[0.3880208432674408,0.2953703701496124,0.526562511920929,0.972222238779068],[0.3421875238418579,0.2212962955236435,0.5369791984558105,0.9712962955236435],[0.26354166865348816,0.04444444179534912,0.5432291924953461,0.9712962508201599]]}]}],"rejected":{"the stone coffin carved into the rock":"count_mismatch"},"rejection_counts":{"count_mismatch":1},"row":{"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_011_scene_001.mp4","scene_key":"--cif7Nv6BI_chunk_011_scene_001","category":"test_vehicles","height":1080,"width":1920,"fps":50.0,"num_frames":1478,"duration":29.560000000000002,"rmot_expressions":[{"expression":"the person wearing jeans who walks towards the stone structure","kind":"motion","is_negative":false,"tracks":[{"timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94],"boxes":[[0.5911458730697632,0.46481481194496155,0.6312500424683094,0.7166666686534882],[0.581250011920929,0.47592592239379883,0.6265625134110451,0.7407407462596893],[0.5807291865348816,0.5361111164093018,0.6244791895151138,0.8287037014961243],[0.5937500596046448,0.5175926089286804,0.637500062584877,0.8166666924953461],[0.5619791746139526,0.5120370388031006,0.6171875111758709,0.8314814865589142],[0.5291666984558105,0.4981481432914734,0.5828125327825546,0.8296296298503876],[0.5453125238418579,0.4842592477798462,0.6010416932404041,0.8388888835906982],[0.5380208492279053,0.49351850152015686,0.5984375178813934,0.8657407164573669],[0.48489585518836975,0.47777777910232544,0.5463541932404041,0.864814817905426],[0.47083336114883423,0.4694444239139557,0.5375000312924385,0.8907407224178314],[0.4635416865348816,0.4749999940395355,0.5395833551883698,0.9342592358589172],[0.4593750238418579,0.41111111640930176,0.544270858168602,0.9000000059604645],[0.4322916865348816,0.43611109256744385,0.5302083566784859,0.9712962508201599],[0.4010416865348816,0.41111111640930176,0.49947918951511383,0.970370352268219],[0.40781253576278687,0.42314815521240234,0.4958333745598793,0.9722222089767456],[0.4411458671092987,0.35555553436279297,0.5338542088866234,0.9731481075286865],[0.41718751192092896,0.35555553436279297,0.5296875163912773,0.9712962508201599],[0.3880208432674408,0.2953703701496124,0.526562511920929,0.972222238779068],[0.3421875238418579,0.2212962955236435,0.5369791984558105,0.9712962955236435],[0.26354166865348816,0.04444444179534912,0.5432291924953461,0.9712962508201599]]}]}],"rmot_expression_count":1,"rmot_negative_count":0,"rmot_motion_count":1}}}}},{"index":16,"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_010_scene_016.mp4","status":"complete","steps":{"decode":{"inputs":{"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_010_scene_016.mp4","sample_fps":4.0},"outputs":{"width":1920,"height":1080,"source_fps":50.0,"duration_seconds":23.22,"frames_kept":90,"blank_frames_dropped":0,"timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88,10.14,10.4,10.66,10.92,11.18,11.44,11.700000000000001,11.96,12.22,12.48,12.74,13.0,13.26,13.52,13.780000000000001,14.040000000000001,14.3,14.56,14.82,15.08,15.34,15.6,15.860000000000001,16.12,16.38,16.64,16.9,17.16,17.42,17.68,17.94,18.2,18.46,18.72,18.98,19.240000000000002,19.5,19.76,20.02,20.28,20.54,20.8,21.06,21.32,21.580000000000002,21.84,22.1,22.36,22.62,22.88,23.14],"error":""}},"propose":{"inputs":{"model":"gemini-3.5-flash-lite","temperature":0.4,"video_bytes":29683518,"system_prompt":"You are annotating video clips to build training data for referring multi-object tracking: given a video and a natural-language description, a model must output the boxes of every object matching that description, over time.\n\nYour job is to write the descriptions. You do NOT output coordinates, boxes, pixel positions, or any numbers about where things are. A separate detector finds the objects.\n\nFor every clip you produce a list of items. Each item has two texts that must NOT be the same:\n\nEXPRESSION — what a person would say to single out the object or objects. It may describe motion, temporal order, interaction with other objects, and spatial relations. It should read like something a person would actually say.\n  Good: \"the black car that's turning left\"\n  Good: \"the cyclist who overtakes the bus\"\n  Good: \"the two people who walk out of frame on the right\"\n\nNOUN_PHRASE — the same object as a plain, timeless visual category with its visible attributes, and NOTHING else. The detector is prompted with this text and looks at ONE frame at a time, so anything about time, order, or action is useless to it and makes it fail.\n  Allowed: category nouns, colours, materials, clothing, size, type\n  Forbidden: verbs and participles of action (\"turning\", \"walks\", \"is moving\"), temporal words (\"then\", \"starts\", \"after\", \"first\", \"last\"), relative clauses (\"that ...\", \"which ...\", \"who ...\"), ordinals and rank words (\"first\", \"leftmost\", \"nearest\"), and anything over six words.\n  \"the black car that's turning left\" -> \"black car\"\n  \"the cyclist who overtakes the bus\" -> \"cyclist\"\n  \"the two people who walk out of frame on the right\" -> \"person\"\n\nKIND — what makes the expression hard: \"motion\", \"appearance\", \"spatial\", or \"interaction\". Aim for about 35% \"motion\" items: motion is the part a single-frame model cannot already solve, so it is the most valuable, but a list of only motion items is not usable training data either.\n\nEXPECTED_COUNT — how many distinct objects in the clip match the EXPRESSION. Count carefully; this is checked against the detector's answer, and a wrong count discards the item.\n\nIS_NEGATIVE — at most ONE item may be a negative, and only when you also write at least one non-negative item. A negative is an expression that recombines attributes that ARE present in the clip into a whole that is NOT. If the only black car is moving, \"the parked black car\" is a good negative. Its NOUN_PHRASE must still be the category that IS present (\"black car\"), because that is what makes the absence checkable rather than merely unverified. A negative's EXPECTED_COUNT is 0.\n\nRules that discard an item, so follow them exactly:\n- Never write coordinates, box values, pixel positions, or percentages of the frame.\n- Never let a NOUN_PHRASE contain a verb of action, a temporal word, a relative clause, or an ordinal.\n- Never write an EXPRESSION whose object is not visible in the clip, unless it is marked IS_NEGATIVE.\n- Prefer objects that are clearly visible and identifiable over background clutter.\n- Do not describe the camera or the shot; describe objects in the scene.\n","user_prompt":"Watch this clip and write between 1 and 3 items following the rules. Cover different objects rather than restating one object several ways, and include any negative in the same list.\n"},"outputs":{"raw_response":"{\n  \"items\": [\n    {\n      \"expression\": \"the man with a beard and black hoodie standing near the stone wall\",\n      \"noun_phrase\": \"man\",\n      \"kind\": \"appearance\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    }\n  ]\n}","raw_items":[{"expression":"the man with a beard and black hoodie standing near the stone wall","noun_phrase":"man","kind":"appearance","expected_count":1,"is_negative":false}],"accepted":[{"expression":"the man with a beard and black hoodie standing near the stone wall","noun_phrase":"man","kind":"appearance","expected_count":1,"is_negative":false}],"gated":[]}},"track":{"inputs":{"tracker":"Sam31MultiplexTracker","model_id":"/mnt/ceph3/ec/jacky/sam3.1","noun_phrases":["man"],"expressions_in":1,"frames":90,"sample_fps":4.0,"frame_size":"1920x1080","timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88,10.14,10.4,10.66,10.92,11.18,11.44,11.700000000000001,11.96,12.22,12.48,12.74,13.0,13.26,13.52,13.780000000000001,14.040000000000001,14.3,14.56,14.82,15.08,15.34,15.6,15.860000000000001,16.12,16.38,16.64,16.9,17.16,17.42,17.68,17.94,18.2,18.46,18.72,18.98,19.240000000000002,19.5,19.76,20.02,20.28,20.54,20.8,21.06,21.32,21.580000000000002,21.84,22.1,22.36,22.62,22.88,23.14],"session":"one session over 90 JPEG frames; then per phrase: reset_session, add_prompt(text) on frame 0, propagate_in_video","passes":1},"outputs":{"raw_tracks":[{"noun_phrase":"man","object_id":0,"observations":[{"timestamp":21.84,"box":[0.8364583849906921,0.40648147463798523,0.9994792342185974,0.9731481373310089],"occluded":false},{"timestamp":22.1,"box":[0.6536458730697632,0.43518516421318054,0.8911458849906921,0.9740740358829498],"occluded":false},{"timestamp":22.36,"box":[0.5666667222976685,0.37962961196899414,0.8322917222976685,0.9759259223937988],"occluded":false},{"timestamp":22.62,"box":[0.483854204416275,0.3583333194255829,0.7734375596046448,0.9759258925914764],"occluded":false},{"timestamp":22.88,"box":[0.42343753576278687,0.3166666626930237,0.7401042282581329,0.9787036776542664],"occluded":false},{"timestamp":23.14,"box":[0.3916666805744171,0.2537037134170532,0.7036458551883698,0.9759259223937988],"occluded":false}],"scores":[0.9532163739204407,0.9532163739204407,0.9532163739204407,0.9532163739204407,0.9532163739204407,0.9532163739204407]}],"tracklets":[{"noun_phrase":"man","object_id":0,"presentation_id":1,"observations":[{"timestamp":21.84,"box":[0.8364583849906921,0.40648147463798523,0.9994792342185974,0.9731481373310089],"occluded":false},{"timestamp":22.1,"box":[0.6536458730697632,0.43518516421318054,0.8911458849906921,0.9740740358829498],"occluded":false},{"timestamp":22.36,"box":[0.5666667222976685,0.37962961196899414,0.8322917222976685,0.9759259223937988],"occluded":false},{"timestamp":22.62,"box":[0.483854204416275,0.3583333194255829,0.7734375596046448,0.9759258925914764],"occluded":false},{"timestamp":22.88,"box":[0.42343753576278687,0.3166666626930237,0.7401042282581329,0.9787036776542664],"occluded":false},{"timestamp":23.14,"box":[0.3916666805744171,0.2537037134170532,0.7036458551883698,0.9759259223937988],"occluded":false}],"mean_score":0.9532163739204407}],"dropped":{},"fates":[{"noun_phrase":"man","object_id":0,"frames":6,"mean_score":0.9532163739204407,"mean_area":0.16542728190701203,"outcome":"kept","presentation_id":1,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null}],"gates":{"min_track_frames":3,"min_mean_score":0.5,"merge_iou":0.8,"max_tracklets":12}}},"render":{"inputs":{"tracklets":1,"as_video":true,"fps":1.0,"max_frames":30},"outputs":{"media":"/home/zzk/data/video_bbox/vlm_gemini3.5_lite/renders/--cif7Nv6BI_chunk_010_scene_016_choices.mp4","media_bytes":7898341,"frames":[{"index":1,"timestamp":0.0,"file":"--cif7Nv6BI_chunk_010_scene_016_choices.mp4"},{"index":2,"timestamp":1.04,"file":"--cif7Nv6BI_chunk_010_scene_016_choices.mp4"},{"index":3,"timestamp":2.08,"file":"--cif7Nv6BI_chunk_010_scene_016_choices.mp4"},{"index":4,"timestamp":3.12,"file":"--cif7Nv6BI_chunk_010_scene_016_choices.mp4"},{"index":5,"timestamp":3.9,"file":"--cif7Nv6BI_chunk_010_scene_016_choices.mp4"},{"index":6,"timestamp":4.94,"file":"--cif7Nv6BI_chunk_010_scene_016_choices.mp4"},{"index":7,"timestamp":5.98,"file":"--cif7Nv6BI_chunk_010_scene_016_choices.mp4"},{"index":8,"timestamp":7.0200000000000005,"file":"--cif7Nv6BI_chunk_010_scene_016_choices.mp4"},{"index":9,"timestamp":8.06,"file":"--cif7Nv6BI_chunk_010_scene_016_choices.mp4"},{"index":10,"timestamp":9.1,"file":"--cif7Nv6BI_chunk_010_scene_016_choices.mp4"},{"index":11,"timestamp":10.14,"file":"--cif7Nv6BI_chunk_010_scene_016_choices.mp4"},{"index":12,"timestamp":11.18,"file":"--cif7Nv6BI_chunk_010_scene_016_choices.mp4"},{"index":13,"timestamp":11.96,"file":"--cif7Nv6BI_chunk_010_scene_016_choices.mp4"},{"index":14,"timestamp":13.0,"file":"--cif7Nv6BI_chunk_010_scene_016_choices.mp4"},{"index":15,"timestamp":14.040000000000001,"file":"--cif7Nv6BI_chunk_010_scene_016_choices.mp4"},{"index":16,"timestamp":15.08,"file":"--cif7Nv6BI_chunk_010_scene_016_choices.mp4"},{"index":17,"timestamp":16.12,"file":"--cif7Nv6BI_chunk_010_scene_016_choices.mp4"},{"index":18,"timestamp":17.16,"file":"--cif7Nv6BI_chunk_010_scene_016_choices.mp4"},{"index":19,"timestamp":18.2,"file":"--cif7Nv6BI_chunk_010_scene_016_choices.mp4"},{"index":20,"timestamp":19.240000000000002,"file":"--cif7Nv6BI_chunk_010_scene_016_choices.mp4"},{"index":21,"timestamp":20.02,"file":"--cif7Nv6BI_chunk_010_scene_016_choices.mp4"},{"index":22,"timestamp":21.06,"file":"--cif7Nv6BI_chunk_010_scene_016_choices.mp4"},{"index":23,"timestamp":22.1,"file":"--cif7Nv6BI_chunk_010_scene_016_choices.mp4"},{"index":24,"timestamp":23.14,"file":"--cif7Nv6BI_chunk_010_scene_016_choices.mp4"}]}},"select":{"inputs":{"model":"gemini-3.5-flash-lite","temperature":0.0,"as_video":true,"media_bytes":7898341,"media":["--cif7Nv6BI_chunk_010_scene_016_choices.mp4"],"choice_frames":24,"system_prompt":"You are labelling video data for referring multi-object tracking.\n\nYou will see a numbered sequence of frames from one clip, in time order. Each frame has its own number and timestamp printed in the top-left corner. Candidate objects have been detected for you and drawn on the frames as coloured boxes, each labelled with a number. The same object keeps the same number and the same colour on every frame. Below each frame is a strip of enlarged crops of those same numbered boxes, so you can identify a small or distant object.\n\nFor each description you are given, answer two questions:\n\n1. MATCHING_IDS — which of the numbered objects the description refers to. List every number that matches, or an EMPTY LIST if none of them do. An empty list is a correct and expected answer: some descriptions are deliberately about something that is not in this clip. Do not pick a number just to have picked one, and do not pick a number for an object that merely resembles the description.\n\n2. START_FRAME and END_FRAME — the frame numbers between which the described state actually holds. If the description is \"the car that's turning left\", the interval is the frames in which that car is turning, NOT every frame in which the car is visible. Use the frame numbers printed on the frames. START_FRAME must not be greater than END_FRAME. When MATCHING_IDS is empty, give the full range (the first and last frame numbers); it is ignored.\n\n3. CONFIDENCE — how sure you are of the MATCHING_IDS answer, from 0.0 to 1.0. Be honest: a low number costs the example nothing, and an overconfident wrong answer becomes a wrong label.\n\nRules:\n- Only use numbers that are actually drawn on the frames.\n- Never output coordinates, box values, or pixel positions.\n- Judge motion by comparing the object's position across frames, not by guessing from one frame.\n","user_prompt":"The video has 24 frames, numbered 1 to 24. Each frame is one moment of the clip and has its number burned into the top left corner.\n\nAnswer for each of the following descriptions, in the order given:\n1. the man with a beard and black hoodie standing near the stone wall","questions":["the man with a beard and black hoodie standing near the stone wall"]},"outputs":{"raw_response":"{\n  \"answers\": [\n    {\n      \"description_index\": 1,\n      \"matching_ids\": [\n        1\n      ],\n      \"start_frame\": 23,\n      \"end_frame\": 24,\n      \"confidence\": 1.0\n    }\n  ]\n}","raw_answers":[{"description_index":1,"matching_ids":[1],"start_frame":23,"end_frame":24,"confidence":1.0}],"answers":[{"expression":"the man with a beard and black hoodie standing near the stone wall","matching_presentation_ids":[1],"start_choice_frame":23,"end_choice_frame":24,"confidence":1.0,"self_reported_confidence":1.0}]}},"verify":{"inputs":{"expressions":1,"tracklets":1,"selections":1,"gates":"interval, count, unknown ids, confidence, then geometry (motion)"},"outputs":{"kept":[{"expression":"the man with a beard and black hoodie standing near the stone wall","kind":"appearance","is_negative":false,"tracks":[{"timestamps":[22.1,22.36,22.62,22.88,23.14],"boxes":[[0.6536458730697632,0.43518516421318054,0.8911458849906921,0.9740740358829498],[0.5666667222976685,0.37962961196899414,0.8322917222976685,0.9759259223937988],[0.483854204416275,0.3583333194255829,0.7734375596046448,0.9759258925914764],[0.42343753576278687,0.3166666626930237,0.7401042282581329,0.9787036776542664],[0.3916666805744171,0.2537037134170532,0.7036458551883698,0.9759259223937988]]}]}],"rejected":{},"rejection_counts":{},"row":{"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_010_scene_016.mp4","scene_key":"--cif7Nv6BI_chunk_010_scene_016","category":"test_vehicles","height":1080,"width":1920,"fps":50.0,"num_frames":1161,"duration":23.22,"rmot_expressions":[{"expression":"the man with a beard and black hoodie standing near the stone wall","kind":"appearance","is_negative":false,"tracks":[{"timestamps":[22.1,22.36,22.62,22.88,23.14],"boxes":[[0.6536458730697632,0.43518516421318054,0.8911458849906921,0.9740740358829498],[0.5666667222976685,0.37962961196899414,0.8322917222976685,0.9759259223937988],[0.483854204416275,0.3583333194255829,0.7734375596046448,0.9759258925914764],[0.42343753576278687,0.3166666626930237,0.7401042282581329,0.9787036776542664],[0.3916666805744171,0.2537037134170532,0.7036458551883698,0.9759259223937988]]}]}],"rmot_expression_count":1,"rmot_negative_count":0,"rmot_motion_count":0}}}}},{"index":17,"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--X6fL8DOMM_chunk_000_scene_003.mp4","status":"complete","steps":{"decode":{"inputs":{"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--X6fL8DOMM_chunk_000_scene_003.mp4","sample_fps":4.0},"outputs":{"width":1920,"height":1080,"source_fps":29.97002997002997,"duration_seconds":16.783433333333335,"frames_kept":63,"blank_frames_dropped":0,"timestamps":[0.0,0.26693333333333336,0.5338666666666667,0.8008000000000001,1.0677333333333334,1.3346666666666667,1.6016000000000001,1.8685333333333334,2.135466666666667,2.4024,2.6693333333333333,2.936266666666667,3.2032000000000003,3.4701333333333335,3.7370666666666668,4.0040000000000004,4.270933333333334,4.537866666666667,4.8048,5.071733333333333,5.338666666666667,5.6056,5.872533333333334,6.139466666666667,6.4064000000000005,6.673333333333334,6.940266666666667,7.2072,7.4741333333333335,7.741066666666667,8.008000000000001,8.274933333333333,8.541866666666667,8.8088,9.075733333333334,9.342666666666666,9.6096,9.876533333333334,10.143466666666667,10.410400000000001,10.677333333333333,10.944266666666667,11.2112,11.478133333333334,11.745066666666668,12.012,12.278933333333335,12.545866666666667,12.812800000000001,13.079733333333333,13.346666666666668,13.6136,13.880533333333334,14.147466666666668,14.4144,14.681333333333335,14.948266666666667,15.215200000000001,15.482133333333334,15.749066666666668,16.016000000000002,16.282933333333332,16.549866666666667],"error":""}},"propose":{"inputs":{"model":"gemini-3.5-flash-lite","temperature":0.4,"video_bytes":21829724,"system_prompt":"You are annotating video clips to build training data for referring multi-object tracking: given a video and a natural-language description, a model must output the boxes of every object matching that description, over time.\n\nYour job is to write the descriptions. You do NOT output coordinates, boxes, pixel positions, or any numbers about where things are. A separate detector finds the objects.\n\nFor every clip you produce a list of items. Each item has two texts that must NOT be the same:\n\nEXPRESSION — what a person would say to single out the object or objects. It may describe motion, temporal order, interaction with other objects, and spatial relations. It should read like something a person would actually say.\n  Good: \"the black car that's turning left\"\n  Good: \"the cyclist who overtakes the bus\"\n  Good: \"the two people who walk out of frame on the right\"\n\nNOUN_PHRASE — the same object as a plain, timeless visual category with its visible attributes, and NOTHING else. The detector is prompted with this text and looks at ONE frame at a time, so anything about time, order, or action is useless to it and makes it fail.\n  Allowed: category nouns, colours, materials, clothing, size, type\n  Forbidden: verbs and participles of action (\"turning\", \"walks\", \"is moving\"), temporal words (\"then\", \"starts\", \"after\", \"first\", \"last\"), relative clauses (\"that ...\", \"which ...\", \"who ...\"), ordinals and rank words (\"first\", \"leftmost\", \"nearest\"), and anything over six words.\n  \"the black car that's turning left\" -> \"black car\"\n  \"the cyclist who overtakes the bus\" -> \"cyclist\"\n  \"the two people who walk out of frame on the right\" -> \"person\"\n\nKIND — what makes the expression hard: \"motion\", \"appearance\", \"spatial\", or \"interaction\". Aim for about 35% \"motion\" items: motion is the part a single-frame model cannot already solve, so it is the most valuable, but a list of only motion items is not usable training data either.\n\nEXPECTED_COUNT — how many distinct objects in the clip match the EXPRESSION. Count carefully; this is checked against the detector's answer, and a wrong count discards the item.\n\nIS_NEGATIVE — at most ONE item may be a negative, and only when you also write at least one non-negative item. A negative is an expression that recombines attributes that ARE present in the clip into a whole that is NOT. If the only black car is moving, \"the parked black car\" is a good negative. Its NOUN_PHRASE must still be the category that IS present (\"black car\"), because that is what makes the absence checkable rather than merely unverified. A negative's EXPECTED_COUNT is 0.\n\nRules that discard an item, so follow them exactly:\n- Never write coordinates, box values, pixel positions, or percentages of the frame.\n- Never let a NOUN_PHRASE contain a verb of action, a temporal word, a relative clause, or an ordinal.\n- Never write an EXPRESSION whose object is not visible in the clip, unless it is marked IS_NEGATIVE.\n- Prefer objects that are clearly visible and identifiable over background clutter.\n- Do not describe the camera or the shot; describe objects in the scene.\n","user_prompt":"Watch this clip and write between 1 and 3 items following the rules. Cover different objects rather than restating one object several ways, and include any negative in the same list.\n"},"outputs":{"raw_response":"{\n  \"items\": [\n    {\n      \"expression\": \"the man with sunglasses sitting in the front passenger seat\",\n      \"noun_phrase\": \"man\",\n      \"kind\": \"spatial\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the young man sitting in the back seat wearing a white t-shirt\",\n      \"noun_phrase\": \"man\",\n      \"kind\": \"spatial\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    }\n  ]\n}","raw_items":[{"expression":"the man with sunglasses sitting in the front passenger seat","noun_phrase":"man","kind":"spatial","expected_count":1,"is_negative":false},{"expression":"the young man sitting in the back seat wearing a white t-shirt","noun_phrase":"man","kind":"spatial","expected_count":1,"is_negative":false}],"accepted":[{"expression":"the man with sunglasses sitting in the front passenger seat","noun_phrase":"man","kind":"spatial","expected_count":1,"is_negative":false},{"expression":"the young man sitting in the back seat wearing a white t-shirt","noun_phrase":"man","kind":"spatial","expected_count":1,"is_negative":false}],"gated":[]}},"track":{"inputs":{"tracker":"Sam31MultiplexTracker","model_id":"/mnt/ceph3/ec/jacky/sam3.1","noun_phrases":["man"],"expressions_in":2,"frames":63,"sample_fps":4.0,"frame_size":"1920x1080","timestamps":[0.0,0.26693333333333336,0.5338666666666667,0.8008000000000001,1.0677333333333334,1.3346666666666667,1.6016000000000001,1.8685333333333334,2.135466666666667,2.4024,2.6693333333333333,2.936266666666667,3.2032000000000003,3.4701333333333335,3.7370666666666668,4.0040000000000004,4.270933333333334,4.537866666666667,4.8048,5.071733333333333,5.338666666666667,5.6056,5.872533333333334,6.139466666666667,6.4064000000000005,6.673333333333334,6.940266666666667,7.2072,7.4741333333333335,7.741066666666667,8.008000000000001,8.274933333333333,8.541866666666667,8.8088,9.075733333333334,9.342666666666666,9.6096,9.876533333333334,10.143466666666667,10.410400000000001,10.677333333333333,10.944266666666667,11.2112,11.478133333333334,11.745066666666668,12.012,12.278933333333335,12.545866666666667,12.812800000000001,13.079733333333333,13.346666666666668,13.6136,13.880533333333334,14.147466666666668,14.4144,14.681333333333335,14.948266666666667,15.215200000000001,15.482133333333334,15.749066666666668,16.016000000000002,16.282933333333332,16.549866666666667],"session":"one session over 63 JPEG frames; then per phrase: reset_session, add_prompt(text) on frame 0, propagate_in_video","passes":1},"outputs":{"raw_tracks":[{"noun_phrase":"man","object_id":0,"observations":[{"timestamp":0.0,"box":[0.6380208730697632,0.10648147761821747,0.9994792342185974,0.9990740269422531],"occluded":false},{"timestamp":0.26693333333333336,"box":[0.6291667222976685,0.11203703284263611,0.9994792342185974,0.9990740716457367],"occluded":false},{"timestamp":0.5338666666666667,"box":[0.6177083849906921,0.15648147463798523,0.9994792342185974,0.9990740716457367],"occluded":false},{"timestamp":0.8008000000000001,"box":[0.6411458849906921,0.1518518477678299,0.9994792342185974,0.9990740567445755],"occluded":false},{"timestamp":1.0677333333333334,"box":[0.6598958969116211,0.1666666567325592,0.9994792342185974,0.9990740716457367],"occluded":false},{"timestamp":1.3346666666666667,"box":[0.518750011920929,0.14629629254341125,0.999479204416275,0.9990740716457367],"occluded":false},{"timestamp":1.6016000000000001,"box":[0.7005208730697632,0.16944444179534912,0.9994792342185974,0.9990740418434143],"occluded":false},{"timestamp":1.8685333333333334,"box":[0.7260417342185974,0.1796296238899231,0.9994792342185974,0.9990740418434143],"occluded":false},{"timestamp":2.135466666666667,"box":[0.7494792342185974,0.17685185372829437,0.9994792342185974,0.9990740865468979],"occluded":false},{"timestamp":2.4024,"box":[0.5604166984558105,0.1648148149251938,0.9994792342185974,0.9990740865468979],"occluded":false},{"timestamp":2.6693333333333333,"box":[0.7864583730697632,0.1379629671573639,0.9994792193174362,0.9990740716457367],"occluded":false},{"timestamp":2.936266666666667,"box":[0.8015625476837158,0.125,0.9994792193174362,0.9990740418434143],"occluded":false},{"timestamp":3.2032000000000003,"box":[0.8213542103767395,0.5370370149612427,0.9994792193174362,0.9990740418434143],"occluded":false},{"timestamp":3.4701333333333335,"box":[0.8291667103767395,0.555555522441864,0.9994792193174362,0.9990740418434143],"occluded":false},{"timestamp":3.7370666666666668,"box":[0.8447917103767395,0.5916666388511658,0.9994792193174362,0.9990740418434143],"occluded":false},{"timestamp":4.0040000000000004,"box":[0.8437500596046448,0.5990740656852722,0.9994792342185974,0.9990740716457367],"occluded":false},{"timestamp":4.270933333333334,"box":[0.8291667103767395,0.2518518567085266,0.9994792193174362,0.9990740418434143],"occluded":false},{"timestamp":4.537866666666667,"box":[0.8338541984558105,0.2666666507720947,0.999479204416275,0.9990740418434143],"occluded":false},{"timestamp":4.8048,"box":[0.7984375357627869,0.17499999701976776,0.9994792193174362,0.9990740269422531],"occluded":false},{"timestamp":5.071733333333333,"box":[0.7880208492279053,0.1379629671573639,0.9994791895151138,0.9990740716457367],"occluded":false},{"timestamp":5.338666666666667,"box":[0.7890625596046448,0.07592592388391495,0.9994792342185974,0.9990740790963173],"occluded":false},{"timestamp":5.6056,"box":[0.7671875357627869,0.05000000074505806,0.9994792193174362,0.9990740306675434],"occluded":false},{"timestamp":5.872533333333334,"box":[0.7484375238418579,0.01666666567325592,0.999479204416275,0.9990740567445755],"occluded":false},{"timestamp":6.139466666666667,"box":[0.7541667222976685,0.0,0.9994792342185974,0.9990740418434143],"occluded":false},{"timestamp":6.4064000000000005,"box":[0.7385417222976685,0.0,0.9994792342185974,0.9990740418434143],"occluded":false},{"timestamp":6.673333333333334,"box":[0.7755208611488342,0.0,0.999479204416275,0.9990740418434143],"occluded":false},{"timestamp":6.940266666666667,"box":[0.7984375357627869,0.06111111119389534,0.9994792193174362,0.9990740604698658],"occluded":false},{"timestamp":7.2072,"box":[0.8281250596046448,0.3314814865589142,0.9994792342185974,0.9990740716457367],"occluded":false},{"timestamp":7.4741333333333335,"box":[0.8526042103767395,0.6444444060325623,0.9994792193174362,0.9990740418434143],"occluded":false},{"timestamp":7.741066666666667,"box":[0.8708333969116211,0.7074074149131775,0.9994792342185974,0.9990740716457367],"occluded":false},{"timestamp":8.008000000000001,"box":[0.8593750596046448,0.6814814805984497,0.9994792342185974,0.9990740716457367],"occluded":false},{"timestamp":8.274933333333333,"box":[0.8473958969116211,0.6037036776542664,0.9994792342185974,0.9990740418434143],"occluded":false},{"timestamp":8.541866666666667,"box":[0.8177083730697632,0.5222222208976746,0.9994792193174362,0.9990740716457367],"occluded":false},{"timestamp":8.8088,"box":[0.8119792342185974,0.18703703582286835,0.9994792491197586,0.9990740865468979],"occluded":false},{"timestamp":9.075733333333334,"box":[0.7984375357627869,0.05092592537403107,0.9994792193174362,0.9990740567445755],"occluded":false},{"timestamp":9.342666666666666,"box":[0.8109375238418579,0.031481482088565826,0.999479204416275,0.9990740790963173],"occluded":false},{"timestamp":9.6096,"box":[0.5328125357627869,0.010185184888541698,0.9994792342185974,0.999074044637382],"occluded":false},{"timestamp":9.876533333333334,"box":[0.4911458492279053,0.03888888657093048,0.9994792342185974,0.9990740567445755],"occluded":false},{"timestamp":10.143466666666667,"box":[0.5328125357627869,0.07129629701375961,0.9994792342185974,0.9990740641951561],"occluded":false},{"timestamp":10.410400000000001,"box":[0.7354167103767395,0.040740739554166794,0.9994792342185974,0.9990740530192852],"occluded":false},{"timestamp":10.677333333333333,"box":[0.4911458492279053,0.014814814552664757,0.9994792342185974,0.999074062332511],"occluded":false},{"timestamp":10.944266666666667,"box":[0.768750011920929,0.028703702613711357,0.9994791895151138,0.9990740548819304],"occluded":false},{"timestamp":11.2112,"box":[0.5744792222976685,0.07222221791744232,0.9994792342185974,0.9990740269422531],"occluded":false},{"timestamp":11.478133333333334,"box":[0.8463541865348816,0.1259259283542633,0.9994791895151138,0.9990740716457367],"occluded":false},{"timestamp":11.745066666666668,"box":[0.8848958611488342,0.6361110806465149,0.9994791969656944,0.9990740418434143],"occluded":false},{"timestamp":12.012,"box":[0.8802083730697632,0.6305555701255798,0.9994792118668556,0.9990740716457367],"occluded":false},{"timestamp":12.278933333333335,"box":[0.8526042103767395,0.17407406866550446,0.9994792193174362,0.9990740567445755],"occluded":false},{"timestamp":12.545866666666667,"box":[0.8442708849906921,0.14722222089767456,0.9994792193174362,0.9990740418434143],"occluded":false},{"timestamp":12.812800000000001,"box":[0.8354167342185974,0.1518518477678299,0.9994792491197586,0.9990740567445755],"occluded":false},{"timestamp":13.079733333333333,"box":[0.8375000357627869,0.1675925850868225,0.999479204416275,0.9990740418434143],"occluded":false},{"timestamp":13.346666666666668,"box":[0.8276042342185974,0.17037037014961243,0.9994792491197586,0.9990740716457367],"occluded":false},{"timestamp":13.6136,"box":[0.8348958492279053,0.182407408952713,0.9994791895151138,0.9990740716457367],"occluded":false},{"timestamp":13.880533333333334,"box":[0.8302083611488342,0.1509259194135666,0.999479204416275,0.9990740269422531],"occluded":false},{"timestamp":14.147466666666668,"box":[0.8203125596046448,0.1240740716457367,0.9994792342185974,0.9990740716457367],"occluded":false},{"timestamp":14.4144,"box":[0.8151041865348816,0.13055555522441864,0.9994791895151138,0.9990740865468979],"occluded":false},{"timestamp":14.681333333333335,"box":[0.7979167103767395,0.14629629254341125,0.9994792193174362,0.9990740716457367],"occluded":false},{"timestamp":14.948266666666667,"box":[0.7901042103767395,0.1398148089647293,0.9994792193174362,0.9990740567445755],"occluded":false},{"timestamp":15.215200000000001,"box":[0.7942708730697632,0.13611111044883728,0.9994792193174362,0.9990740716457367],"occluded":false},{"timestamp":15.482133333333334,"box":[0.7859375476837158,0.13240741193294525,0.9994792193174362,0.9990740865468979],"occluded":false},{"timestamp":15.749066666666668,"box":[0.7718750238418579,0.1259259283542633,0.999479204416275,0.9990740716457367],"occluded":false},{"timestamp":16.016000000000002,"box":[0.7614583969116211,0.12314814329147339,0.9994792491197586,0.9990740418434143],"occluded":false},{"timestamp":16.282933333333332,"box":[0.7473958730697632,0.10185185074806213,0.9994792342185974,0.9990740716457367],"occluded":false},{"timestamp":16.549866666666667,"box":[0.7234375476837158,0.10740740597248077,0.9994792342185974,0.9990740567445755],"occluded":false}],"scores":[0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875,0.96875]},{"noun_phrase":"man","object_id":1,"observations":[{"timestamp":0.0,"box":[0.0,0.0,0.4791666865348816,0.9990740418434143],"occluded":false},{"timestamp":0.26693333333333336,"box":[0.0,0.0,0.47708335518836975,0.9990740418434143],"occluded":false},{"timestamp":0.5338666666666667,"box":[0.0,0.0,0.44583335518836975,0.9990740418434143],"occluded":false},{"timestamp":0.8008000000000001,"box":[0.0,0.0,0.4583333432674408,0.9990740418434143],"occluded":false},{"timestamp":1.0677333333333334,"box":[0.0,0.0,0.476041704416275,0.9990740418434143],"occluded":false},{"timestamp":1.3346666666666667,"box":[0.0,0.0,0.4833333492279053,0.9990740418434143],"occluded":false},{"timestamp":1.6016000000000001,"box":[0.0,0.0,0.5005208849906921,0.9990740418434143],"occluded":false},{"timestamp":1.8685333333333334,"box":[0.0,0.0,0.5104166865348816,0.9990740418434143],"occluded":false},{"timestamp":2.135466666666667,"box":[0.0,0.0,0.5223958492279053,0.9990740418434143],"occluded":false},{"timestamp":2.4024,"box":[0.0,0.010185184888541698,0.47447919845581055,0.999074044637382],"occluded":false},{"timestamp":2.6693333333333333,"box":[0.0,0.0,0.5213541984558105,0.9990740418434143],"occluded":false},{"timestamp":2.936266666666667,"box":[0.0,0.0,0.542187511920929,0.9990740418434143],"occluded":false},{"timestamp":3.2032000000000003,"box":[0.0,0.0,0.5562500357627869,0.9990740418434143],"occluded":false},{"timestamp":3.4701333333333335,"box":[0.0,0.0,0.5619791746139526,0.9990740418434143],"occluded":false},{"timestamp":3.7370666666666668,"box":[0.0,0.0,0.5687500238418579,0.9990740418434143],"occluded":false},{"timestamp":4.0040000000000004,"box":[0.0,0.0,0.5770833492279053,0.9990740418434143],"occluded":false},{"timestamp":4.270933333333334,"box":[0.0,0.0,0.5692708492279053,0.9972221851348877],"occluded":false},{"timestamp":4.537866666666667,"box":[0.0,0.0,0.5864583849906921,0.9990740418434143],"occluded":false},{"timestamp":4.8048,"box":[0.0,0.0,0.5739583373069763,0.9990740418434143],"occluded":false},{"timestamp":5.071733333333333,"box":[0.0,0.0,0.5901042222976685,0.9990740418434143],"occluded":false},{"timestamp":5.338666666666667,"box":[0.0,0.0,0.6276041865348816,0.9990740418434143],"occluded":false},{"timestamp":5.6056,"box":[0.0,0.0,0.6062500476837158,0.9990740418434143],"occluded":false},{"timestamp":5.872533333333334,"box":[0.0,0.0,0.5911458730697632,0.9990740418434143],"occluded":false},{"timestamp":6.139466666666667,"box":[0.0,0.0,0.5963541865348816,0.9990740418434143],"occluded":false},{"timestamp":6.4064000000000005,"box":[0.0,0.0,0.5661458373069763,0.9990740418434143],"occluded":false},{"timestamp":6.673333333333334,"box":[0.0,0.0,0.5703125,0.9990740418434143],"occluded":false},{"timestamp":6.940266666666667,"box":[0.0,0.0,0.5807291865348816,0.9990740418434143],"occluded":false},{"timestamp":7.2072,"box":[0.0,0.0,0.5963541865348816,0.9990740418434143],"occluded":false},{"timestamp":7.4741333333333335,"box":[0.0,0.0,0.9078125357627869,0.9990740418434143],"occluded":false},{"timestamp":7.741066666666667,"box":[0.0,0.0,0.9218750596046448,0.9990740418434143],"occluded":false},{"timestamp":8.008000000000001,"box":[0.0,0.0,0.9494792222976685,0.9990740418434143],"occluded":false},{"timestamp":8.274933333333333,"box":[0.0,0.0,0.5869792103767395,0.9990740418434143],"occluded":false},{"timestamp":8.541866666666667,"box":[0.0,0.0,0.5505208373069763,0.9990740418434143],"occluded":false},{"timestamp":8.8088,"box":[0.0,0.0,0.5729166865348816,0.9990740418434143],"occluded":false},{"timestamp":9.075733333333334,"box":[0.0,0.0,0.5531250238418579,0.9990740418434143],"occluded":false},{"timestamp":9.342666666666666,"box":[0.01927083358168602,0.0,0.5750000439584255,0.9990740418434143],"occluded":false},{"timestamp":9.6096,"box":[0.0,0.0,0.5786458849906921,0.9990740418434143],"occluded":false},{"timestamp":9.876533333333334,"box":[0.0,0.0,0.5843750238418579,0.9990740418434143],"occluded":false},{"timestamp":10.143466666666667,"box":[0.0,0.0,0.5723958611488342,0.9990740418434143],"occluded":false},{"timestamp":10.410400000000001,"box":[0.0,0.0,0.5473958849906921,0.9990740418434143],"occluded":false},{"timestamp":10.677333333333333,"box":[0.0,0.0,0.5666667222976685,0.9990740418434143],"occluded":false},{"timestamp":10.944266666666667,"box":[0.0,0.0,0.5625,0.9990740418434143],"occluded":false},{"timestamp":11.2112,"box":[0.0,0.0,0.589062511920929,0.9990740418434143],"occluded":false},{"timestamp":11.478133333333334,"box":[0.0,0.0,0.6145833730697632,0.9990740418434143],"occluded":false},{"timestamp":11.745066666666668,"box":[0.0,0.0,0.6369792222976685,0.9990740418434143],"occluded":false},{"timestamp":12.012,"box":[0.0,0.0,0.6427083611488342,0.9990740418434143],"occluded":false},{"timestamp":12.278933333333335,"box":[0.0,0.0,0.6312500238418579,0.9990740418434143],"occluded":false},{"timestamp":12.545866666666667,"box":[0.0,0.0,0.6333333849906921,0.9990740418434143],"occluded":false},{"timestamp":12.812800000000001,"box":[0.0,0.0,0.6260417103767395,0.9972221851348877],"occluded":false},{"timestamp":13.079733333333333,"box":[0.0,0.0,0.6229166984558105,0.9990740418434143],"occluded":false},{"timestamp":13.346666666666668,"box":[0.0,0.0,0.6312500238418579,0.9990740418434143],"occluded":false},{"timestamp":13.6136,"box":[0.0,0.0,0.6348958611488342,0.9990740418434143],"occluded":false},{"timestamp":13.880533333333334,"box":[0.0,0.0,0.6317708492279053,0.9990740418434143],"occluded":false},{"timestamp":14.147466666666668,"box":[0.0,0.0,0.6302083730697632,0.9990740418434143],"occluded":false},{"timestamp":14.4144,"box":[0.0,0.0,0.6072916984558105,0.9990740418434143],"occluded":false},{"timestamp":14.681333333333335,"box":[0.0,0.0,0.5864583849906921,0.9990740418434143],"occluded":false},{"timestamp":14.948266666666667,"box":[0.0,0.0,0.5859375596046448,0.9990740418434143],"occluded":false},{"timestamp":15.215200000000001,"box":[0.0,0.0,0.5921875238418579,0.9990740418434143],"occluded":false},{"timestamp":15.482133333333334,"box":[0.0,0.0,0.5869792103767395,0.9990740418434143],"occluded":false},{"timestamp":15.749066666666668,"box":[0.0,0.0,0.5828125476837158,0.9990740418434143],"occluded":false},{"timestamp":16.016000000000002,"box":[0.0,0.0,0.5744792222976685,0.9990740418434143],"occluded":false},{"timestamp":16.282933333333332,"box":[0.0,0.0,0.5708333849906921,0.9990740418434143],"occluded":false},{"timestamp":16.549866666666667,"box":[0.0026041667442768812,0.0,0.5437500278931111,0.9990740418434143],"occluded":false}],"scores":[0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562,0.9609756469726562]},{"noun_phrase":"man","object_id":2,"observations":[{"timestamp":0.0,"box":[0.3281250298023224,0.24537037312984467,0.5427083671092987,0.9907407015562057],"occluded":false},{"timestamp":0.26693333333333336,"box":[0.33125001192092896,0.22962962090969086,0.5505208522081375,0.9990740269422531],"occluded":false},{"timestamp":0.5338666666666667,"box":[0.32864585518836975,0.26851850748062134,0.5328125357627869,0.9990740418434143],"occluded":false},{"timestamp":0.8008000000000001,"box":[0.34322917461395264,0.2787036895751953,0.5515625178813934,0.9990740418434143],"occluded":false},{"timestamp":1.0677333333333334,"box":[0.35468751192092896,0.2787036895751953,0.5661458522081375,0.9990740418434143],"occluded":false},{"timestamp":1.3346666666666667,"box":[0.3604166805744171,0.25462961196899414,0.5713541954755783,0.9990740418434143],"occluded":false},{"timestamp":1.6016000000000001,"box":[0.3854166865348816,0.27129629254341125,0.5937500298023224,0.9990740716457367],"occluded":false},{"timestamp":1.8685333333333334,"box":[0.39791667461395264,0.287962943315506,0.6098958551883698,0.9990740120410919],"occluded":false},{"timestamp":2.135466666666667,"box":[0.4088541865348816,0.2898148000240326,0.622395858168602,0.9990740716457367],"occluded":false},{"timestamp":2.4024,"box":[0.3973958492279053,0.2916666567325592,0.5973958671092987,0.9990740716457367],"occluded":false},{"timestamp":2.6693333333333333,"box":[0.45104169845581055,0.2611111104488373,0.6598958820104599,0.9990740716457367],"occluded":false},{"timestamp":2.936266666666667,"box":[0.4739583432674408,0.24351851642131805,0.6781250238418579,0.9990740865468979],"occluded":false},{"timestamp":3.2032000000000003,"box":[0.4895833730697632,0.2351851761341095,0.6921875476837158,0.9990740716457367],"occluded":false},{"timestamp":3.4701333333333335,"box":[0.4989583492279053,0.2620370388031006,0.6994791924953461,0.9990740418434143],"occluded":false},{"timestamp":3.7370666666666668,"box":[0.5036458373069763,0.28611111640930176,0.7093750089406967,0.9990740418434143],"occluded":false},{"timestamp":4.0040000000000004,"box":[0.5026041865348816,0.2916666567325592,0.7130208611488342,0.9990740716457367],"occluded":false},{"timestamp":4.270933333333334,"box":[0.4932291805744171,0.2898148000240326,0.6984375268220901,0.9972222149372101],"occluded":false},{"timestamp":4.537866666666667,"box":[0.49375003576278687,0.2944444417953491,0.7041667103767395,0.9990740418434143],"occluded":false},{"timestamp":4.8048,"box":[0.46250003576278687,0.2953703701496124,0.6755208820104599,0.9990740716457367],"occluded":false},{"timestamp":5.071733333333333,"box":[0.44062501192092896,0.2657407224178314,0.669270858168602,0.9990740716457367],"occluded":false},{"timestamp":5.338666666666667,"box":[0.42500001192092896,0.23240740597248077,0.6760416924953461,0.9990740567445755],"occluded":false},{"timestamp":5.6056,"box":[0.3984375298023224,0.21296295523643494,0.658854216337204,0.9990740716457367],"occluded":false},{"timestamp":5.872533333333334,"box":[0.38489586114883423,0.18888889253139496,0.6484375298023224,0.9990740865468979],"occluded":false},{"timestamp":6.139466666666667,"box":[0.3671875298023224,0.1814814805984497,0.6541666984558105,0.9990740418434143],"occluded":false},{"timestamp":6.4064000000000005,"box":[0.3411458432674408,0.1657407432794571,0.632291704416275,0.9990740567445755],"occluded":false},{"timestamp":6.673333333333334,"box":[0.3973958492279053,0.21759258210659027,0.6567708551883698,0.9990740269422531],"occluded":false},{"timestamp":6.940266666666667,"box":[0.42239585518836975,0.27037036418914795,0.6718750298023224,0.9990740418434143],"occluded":false},{"timestamp":7.2072,"box":[0.43906253576278687,0.31111109256744385,0.6911458969116211,0.9990740418434143],"occluded":false},{"timestamp":7.4741333333333335,"box":[0.460416704416275,0.3629629611968994,0.6968750506639481,0.9990740418434143],"occluded":false},{"timestamp":7.741066666666667,"box":[0.4635416865348816,0.41018518805503845,0.695833370089531,0.9990740716457367],"occluded":false},{"timestamp":8.008000000000001,"box":[0.45729169249534607,0.4037036895751953,0.688020870089531,0.9990740418434143],"occluded":false},{"timestamp":8.274933333333333,"box":[0.46302086114883423,0.3712962865829468,0.6921875327825546,0.9990740418434143],"occluded":false},{"timestamp":8.541866666666667,"box":[0.4437500238418579,0.37870368361473083,0.6656250357627869,0.9870370328426361],"occluded":false},{"timestamp":8.8088,"box":[0.4515625238418579,0.38240739703178406,0.6890625357627869,0.9861110746860504],"occluded":false},{"timestamp":9.075733333333334,"box":[0.44843751192092896,0.3444444239139557,0.6812500208616257,0.9990740716457367],"occluded":false},{"timestamp":9.342666666666666,"box":[0.46927085518836975,0.3296296298503876,0.7229166924953461,0.9990740716457367],"occluded":false},{"timestamp":9.6096,"box":[0.46562501788139343,0.3185185194015503,0.7239583730697632,0.9990740418434143],"occluded":false},{"timestamp":9.876533333333334,"box":[0.4531250298023224,0.33703702688217163,0.7156250476837158,0.9990740418434143],"occluded":false},{"timestamp":10.143466666666667,"box":[0.43020835518836975,0.34074074029922485,0.6864583790302277,0.9990740418434143],"occluded":false},{"timestamp":10.410400000000001,"box":[0.41041669249534607,0.3212963044643402,0.6682291924953461,0.9990740716457367],"occluded":false},{"timestamp":10.677333333333333,"box":[0.4348958432674408,0.3046296238899231,0.6958333551883698,0.9990740418434143],"occluded":false},{"timestamp":10.944266666666667,"box":[0.44583335518836975,0.3027777671813965,0.7046875357627869,0.9990740418434143],"occluded":false},{"timestamp":11.2112,"box":[0.4750000238418579,0.3212963044643402,0.7276042103767395,0.9990740716457367],"occluded":false},{"timestamp":11.478133333333334,"box":[0.5078125,0.3314814865589142,0.765625,0.9990740716457367],"occluded":false},{"timestamp":11.745066666666668,"box":[0.5385416746139526,0.34351852536201477,0.7953125238418579,0.9990740716457367],"occluded":false},{"timestamp":12.012,"box":[0.5348958373069763,0.32592591643333435,0.792187511920929,0.9990740716457367],"occluded":false},{"timestamp":12.278933333333335,"box":[0.510937511920929,0.3083333373069763,0.7661458551883698,0.9990740418434143],"occluded":false},{"timestamp":12.545866666666667,"box":[0.5057291984558105,0.2944444417953491,0.7588542103767395,0.9990740418434143],"occluded":false},{"timestamp":12.812800000000001,"box":[0.4984375238418579,0.29814815521240234,0.7468750327825546,0.9972222447395325],"occluded":false},{"timestamp":13.079733333333333,"box":[0.49635419249534607,0.3055555522441864,0.7447917014360428,0.9990740716457367],"occluded":false},{"timestamp":13.346666666666668,"box":[0.491666704416275,0.29814815521240234,0.7375000566244125,0.9990740418434143],"occluded":false},{"timestamp":13.6136,"box":[0.49270835518836975,0.3046296238899231,0.741666704416275,0.9990740418434143],"occluded":false},{"timestamp":13.880533333333334,"box":[0.4869791865348816,0.29722222685813904,0.7385416924953461,0.9990740716457367],"occluded":false},{"timestamp":14.147466666666668,"box":[0.48229169845581055,0.2898148000240326,0.7343750596046448,0.9990740716457367],"occluded":false},{"timestamp":14.4144,"box":[0.4776041805744171,0.3046296238899231,0.7244791984558105,0.9990740418434143],"occluded":false},{"timestamp":14.681333333333335,"box":[0.46562501788139343,0.31203702092170715,0.7020833641290665,0.9990740716457367],"occluded":false},{"timestamp":14.948266666666667,"box":[0.4645833671092987,0.3064814805984497,0.7036458849906921,0.9990740418434143],"occluded":false},{"timestamp":15.215200000000001,"box":[0.46250003576278687,0.3166666626930237,0.7052083760499954,0.9990740418434143],"occluded":false},{"timestamp":15.482133333333334,"box":[0.4531250298023224,0.31296294927597046,0.6989583820104599,0.9990740418434143],"occluded":false},{"timestamp":15.749066666666668,"box":[0.44322919845581055,0.31018519401550293,0.6911458820104599,0.9990740418434143],"occluded":false},{"timestamp":16.016000000000002,"box":[0.4322916865348816,0.31388887763023376,0.6812500357627869,0.9990740716457367],"occluded":false},{"timestamp":16.282933333333332,"box":[0.41979169845581055,0.31111109256744385,0.6703125536441803,0.9990740418434143],"occluded":false},{"timestamp":16.549866666666667,"box":[0.3958333432674408,0.3157407343387604,0.6447916924953461,0.9990740716457367],"occluded":false}],"scores":[0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708]}],"tracklets":[{"noun_phrase":"man","object_id":1,"presentation_id":1,"observations":[{"timestamp":0.0,"box":[0.0,0.0,0.4791666865348816,0.9990740418434143],"occluded":false},{"timestamp":0.26693333333333336,"box":[0.0,0.0,0.47708335518836975,0.9990740418434143],"occluded":false},{"timestamp":0.5338666666666667,"box":[0.0,0.0,0.44583335518836975,0.9990740418434143],"occluded":false},{"timestamp":0.8008000000000001,"box":[0.0,0.0,0.4583333432674408,0.9990740418434143],"occluded":false},{"timestamp":1.0677333333333334,"box":[0.0,0.0,0.476041704416275,0.9990740418434143],"occluded":false},{"timestamp":1.3346666666666667,"box":[0.0,0.0,0.4833333492279053,0.9990740418434143],"occluded":false},{"timestamp":1.6016000000000001,"box":[0.0,0.0,0.5005208849906921,0.9990740418434143],"occluded":false},{"timestamp":1.8685333333333334,"box":[0.0,0.0,0.5104166865348816,0.9990740418434143],"occluded":false},{"timestamp":2.135466666666667,"box":[0.0,0.0,0.5223958492279053,0.9990740418434143],"occluded":false},{"timestamp":2.4024,"box":[0.0,0.010185184888541698,0.47447919845581055,0.999074044637382],"occluded":false},{"timestamp":2.6693333333333333,"box":[0.0,0.0,0.5213541984558105,0.9990740418434143],"occluded":false},{"timestamp":2.936266666666667,"box":[0.0,0.0,0.542187511920929,0.9990740418434143],"occluded":false},{"timestamp":3.2032000000000003,"box":[0.0,0.0,0.5562500357627869,0.9990740418434143],"occluded":false},{"timestamp":3.4701333333333335,"box":[0.0,0.0,0.5619791746139526,0.9990740418434143],"occluded":false},{"timestamp":3.7370666666666668,"box":[0.0,0.0,0.5687500238418579,0.9990740418434143],"occluded":false},{"timestamp":4.0040000000000004,"box":[0.0,0.0,0.5770833492279053,0.9990740418434143],"occluded":false},{"timestamp":4.270933333333334,"box":[0.0,0.0,0.5692708492279053,0.9972221851348877],"occluded":false},{"timestamp":4.537866666666667,"box":[0.0,0.0,0.5864583849906921,0.9990740418434143],"occluded":false},{"timestamp":4.8048,"box":[0.0,0.0,0.5739583373069763,0.9990740418434143],"occluded":false},{"timestamp":5.071733333333333,"box":[0.0,0.0,0.5901042222976685,0.9990740418434143],"occluded":false},{"timestamp":5.338666666666667,"box":[0.0,0.0,0.6276041865348816,0.9990740418434143],"occluded":false},{"timestamp":5.6056,"box":[0.0,0.0,0.6062500476837158,0.9990740418434143],"occluded":false},{"timestamp":5.872533333333334,"box":[0.0,0.0,0.5911458730697632,0.9990740418434143],"occluded":false},{"timestamp":6.139466666666667,"box":[0.0,0.0,0.5963541865348816,0.9990740418434143],"occluded":false},{"timestamp":6.4064000000000005,"box":[0.0,0.0,0.5661458373069763,0.9990740418434143],"occluded":false},{"timestamp":6.673333333333334,"box":[0.0,0.0,0.5703125,0.9990740418434143],"occluded":false},{"timestamp":6.940266666666667,"box":[0.0,0.0,0.5807291865348816,0.9990740418434143],"occluded":false},{"timestamp":7.2072,"box":[0.0,0.0,0.5963541865348816,0.9990740418434143],"occluded":false},{"timestamp":7.4741333333333335,"box":[0.0,0.0,0.9078125357627869,0.9990740418434143],"occluded":false},{"timestamp":7.741066666666667,"box":[0.0,0.0,0.9218750596046448,0.9990740418434143],"occluded":false},{"timestamp":8.008000000000001,"box":[0.0,0.0,0.9494792222976685,0.9990740418434143],"occluded":false},{"timestamp":8.274933333333333,"box":[0.0,0.0,0.5869792103767395,0.9990740418434143],"occluded":false},{"timestamp":8.541866666666667,"box":[0.0,0.0,0.5505208373069763,0.9990740418434143],"occluded":false},{"timestamp":8.8088,"box":[0.0,0.0,0.5729166865348816,0.9990740418434143],"occluded":false},{"timestamp":9.075733333333334,"box":[0.0,0.0,0.5531250238418579,0.9990740418434143],"occluded":false},{"timestamp":9.342666666666666,"box":[0.01927083358168602,0.0,0.5750000439584255,0.9990740418434143],"occluded":false},{"timestamp":9.6096,"box":[0.0,0.0,0.5786458849906921,0.9990740418434143],"occluded":false},{"timestamp":9.876533333333334,"box":[0.0,0.0,0.5843750238418579,0.9990740418434143],"occluded":false},{"timestamp":10.143466666666667,"box":[0.0,0.0,0.5723958611488342,0.9990740418434143],"occluded":false},{"timestamp":10.410400000000001,"box":[0.0,0.0,0.5473958849906921,0.9990740418434143],"occluded":false},{"timestamp":10.677333333333333,"box":[0.0,0.0,0.5666667222976685,0.9990740418434143],"occluded":false},{"timestamp":10.944266666666667,"box":[0.0,0.0,0.5625,0.9990740418434143],"occluded":false},{"timestamp":11.2112,"box":[0.0,0.0,0.589062511920929,0.9990740418434143],"occluded":false},{"timestamp":11.478133333333334,"box":[0.0,0.0,0.6145833730697632,0.9990740418434143],"occluded":false},{"timestamp":11.745066666666668,"box":[0.0,0.0,0.6369792222976685,0.9990740418434143],"occluded":false},{"timestamp":12.012,"box":[0.0,0.0,0.6427083611488342,0.9990740418434143],"occluded":false},{"timestamp":12.278933333333335,"box":[0.0,0.0,0.6312500238418579,0.9990740418434143],"occluded":false},{"timestamp":12.545866666666667,"box":[0.0,0.0,0.6333333849906921,0.9990740418434143],"occluded":false},{"timestamp":12.812800000000001,"box":[0.0,0.0,0.6260417103767395,0.9972221851348877],"occluded":false},{"timestamp":13.079733333333333,"box":[0.0,0.0,0.6229166984558105,0.9990740418434143],"occluded":false},{"timestamp":13.346666666666668,"box":[0.0,0.0,0.6312500238418579,0.9990740418434143],"occluded":false},{"timestamp":13.6136,"box":[0.0,0.0,0.6348958611488342,0.9990740418434143],"occluded":false},{"timestamp":13.880533333333334,"box":[0.0,0.0,0.6317708492279053,0.9990740418434143],"occluded":false},{"timestamp":14.147466666666668,"box":[0.0,0.0,0.6302083730697632,0.9990740418434143],"occluded":false},{"timestamp":14.4144,"box":[0.0,0.0,0.6072916984558105,0.9990740418434143],"occluded":false},{"timestamp":14.681333333333335,"box":[0.0,0.0,0.5864583849906921,0.9990740418434143],"occluded":false},{"timestamp":14.948266666666667,"box":[0.0,0.0,0.5859375596046448,0.9990740418434143],"occluded":false},{"timestamp":15.215200000000001,"box":[0.0,0.0,0.5921875238418579,0.9990740418434143],"occluded":false},{"timestamp":15.482133333333334,"box":[0.0,0.0,0.5869792103767395,0.9990740418434143],"occluded":false},{"timestamp":15.749066666666668,"box":[0.0,0.0,0.5828125476837158,0.9990740418434143],"occluded":false},{"timestamp":16.016000000000002,"box":[0.0,0.0,0.5744792222976685,0.9990740418434143],"occluded":false},{"timestamp":16.282933333333332,"box":[0.0,0.0,0.5708333849906921,0.9990740418434143],"occluded":false},{"timestamp":16.549866666666667,"box":[0.0026041667442768812,0.0,0.5437500278931111,0.9990740418434143],"occluded":false}],"mean_score":0.9609756469726562},{"noun_phrase":"man","object_id":0,"presentation_id":2,"observations":[{"timestamp":0.0,"box":[0.6380208730697632,0.10648147761821747,0.9994792342185974,0.9990740269422531],"occluded":false},{"timestamp":0.26693333333333336,"box":[0.6291667222976685,0.11203703284263611,0.9994792342185974,0.9990740716457367],"occluded":false},{"timestamp":0.5338666666666667,"box":[0.6177083849906921,0.15648147463798523,0.9994792342185974,0.9990740716457367],"occluded":false},{"timestamp":0.8008000000000001,"box":[0.6411458849906921,0.1518518477678299,0.9994792342185974,0.9990740567445755],"occluded":false},{"timestamp":1.0677333333333334,"box":[0.6598958969116211,0.1666666567325592,0.9994792342185974,0.9990740716457367],"occluded":false},{"timestamp":1.3346666666666667,"box":[0.518750011920929,0.14629629254341125,0.999479204416275,0.9990740716457367],"occluded":false},{"timestamp":1.6016000000000001,"box":[0.7005208730697632,0.16944444179534912,0.9994792342185974,0.9990740418434143],"occluded":false},{"timestamp":1.8685333333333334,"box":[0.7260417342185974,0.1796296238899231,0.9994792342185974,0.9990740418434143],"occluded":false},{"timestamp":2.135466666666667,"box":[0.7494792342185974,0.17685185372829437,0.9994792342185974,0.9990740865468979],"occluded":false},{"timestamp":2.4024,"box":[0.5604166984558105,0.1648148149251938,0.9994792342185974,0.9990740865468979],"occluded":false},{"timestamp":2.6693333333333333,"box":[0.7864583730697632,0.1379629671573639,0.9994792193174362,0.9990740716457367],"occluded":false},{"timestamp":2.936266666666667,"box":[0.8015625476837158,0.125,0.9994792193174362,0.9990740418434143],"occluded":false},{"timestamp":3.2032000000000003,"box":[0.8213542103767395,0.5370370149612427,0.9994792193174362,0.9990740418434143],"occluded":false},{"timestamp":3.4701333333333335,"box":[0.8291667103767395,0.555555522441864,0.9994792193174362,0.9990740418434143],"occluded":false},{"timestamp":3.7370666666666668,"box":[0.8447917103767395,0.5916666388511658,0.9994792193174362,0.9990740418434143],"occluded":false},{"timestamp":4.0040000000000004,"box":[0.8437500596046448,0.5990740656852722,0.9994792342185974,0.9990740716457367],"occluded":false},{"timestamp":4.270933333333334,"box":[0.8291667103767395,0.2518518567085266,0.9994792193174362,0.9990740418434143],"occluded":false},{"timestamp":4.537866666666667,"box":[0.8338541984558105,0.2666666507720947,0.999479204416275,0.9990740418434143],"occluded":false},{"timestamp":4.8048,"box":[0.7984375357627869,0.17499999701976776,0.9994792193174362,0.9990740269422531],"occluded":false},{"timestamp":5.071733333333333,"box":[0.7880208492279053,0.1379629671573639,0.9994791895151138,0.9990740716457367],"occluded":false},{"timestamp":5.338666666666667,"box":[0.7890625596046448,0.07592592388391495,0.9994792342185974,0.9990740790963173],"occluded":false},{"timestamp":5.6056,"box":[0.7671875357627869,0.05000000074505806,0.9994792193174362,0.9990740306675434],"occluded":false},{"timestamp":5.872533333333334,"box":[0.7484375238418579,0.01666666567325592,0.999479204416275,0.9990740567445755],"occluded":false},{"timestamp":6.139466666666667,"box":[0.7541667222976685,0.0,0.9994792342185974,0.9990740418434143],"occluded":false},{"timestamp":6.4064000000000005,"box":[0.7385417222976685,0.0,0.9994792342185974,0.9990740418434143],"occluded":false},{"timestamp":6.673333333333334,"box":[0.7755208611488342,0.0,0.999479204416275,0.9990740418434143],"occluded":false},{"timestamp":6.940266666666667,"box":[0.7984375357627869,0.06111111119389534,0.9994792193174362,0.9990740604698658],"occluded":false},{"timestamp":7.2072,"box":[0.8281250596046448,0.3314814865589142,0.9994792342185974,0.9990740716457367],"occluded":false},{"timestamp":7.4741333333333335,"box":[0.8526042103767395,0.6444444060325623,0.9994792193174362,0.9990740418434143],"occluded":false},{"timestamp":7.741066666666667,"box":[0.8708333969116211,0.7074074149131775,0.9994792342185974,0.9990740716457367],"occluded":false},{"timestamp":8.008000000000001,"box":[0.8593750596046448,0.6814814805984497,0.9994792342185974,0.9990740716457367],"occluded":false},{"timestamp":8.274933333333333,"box":[0.8473958969116211,0.6037036776542664,0.9994792342185974,0.9990740418434143],"occluded":false},{"timestamp":8.541866666666667,"box":[0.8177083730697632,0.5222222208976746,0.9994792193174362,0.9990740716457367],"occluded":false},{"timestamp":8.8088,"box":[0.8119792342185974,0.18703703582286835,0.9994792491197586,0.9990740865468979],"occluded":false},{"timestamp":9.075733333333334,"box":[0.7984375357627869,0.05092592537403107,0.9994792193174362,0.9990740567445755],"occluded":false},{"timestamp":9.342666666666666,"box":[0.8109375238418579,0.031481482088565826,0.999479204416275,0.9990740790963173],"occluded":false},{"timestamp":9.6096,"box":[0.5328125357627869,0.010185184888541698,0.9994792342185974,0.999074044637382],"occluded":false},{"timestamp":9.876533333333334,"box":[0.4911458492279053,0.03888888657093048,0.9994792342185974,0.9990740567445755],"occluded":false},{"timestamp":10.143466666666667,"box":[0.5328125357627869,0.07129629701375961,0.9994792342185974,0.9990740641951561],"occluded":false},{"timestamp":10.410400000000001,"box":[0.7354167103767395,0.040740739554166794,0.9994792342185974,0.9990740530192852],"occluded":false},{"timestamp":10.677333333333333,"box":[0.4911458492279053,0.014814814552664757,0.9994792342185974,0.999074062332511],"occluded":false},{"timestamp":10.944266666666667,"box":[0.768750011920929,0.028703702613711357,0.9994791895151138,0.9990740548819304],"occluded":false},{"timestamp":11.2112,"box":[0.5744792222976685,0.07222221791744232,0.9994792342185974,0.9990740269422531],"occluded":false},{"timestamp":11.478133333333334,"box":[0.8463541865348816,0.1259259283542633,0.9994791895151138,0.9990740716457367],"occluded":false},{"timestamp":11.745066666666668,"box":[0.8848958611488342,0.6361110806465149,0.9994791969656944,0.9990740418434143],"occluded":false},{"timestamp":12.012,"box":[0.8802083730697632,0.6305555701255798,0.9994792118668556,0.9990740716457367],"occluded":false},{"timestamp":12.278933333333335,"box":[0.8526042103767395,0.17407406866550446,0.9994792193174362,0.9990740567445755],"occluded":false},{"timestamp":12.545866666666667,"box":[0.8442708849906921,0.14722222089767456,0.9994792193174362,0.9990740418434143],"occluded":false},{"timestamp":12.812800000000001,"box":[0.8354167342185974,0.1518518477678299,0.9994792491197586,0.9990740567445755],"occluded":false},{"timestamp":13.079733333333333,"box":[0.8375000357627869,0.1675925850868225,0.999479204416275,0.9990740418434143],"occluded":false},{"timestamp":13.346666666666668,"box":[0.8276042342185974,0.17037037014961243,0.9994792491197586,0.9990740716457367],"occluded":false},{"timestamp":13.6136,"box":[0.8348958492279053,0.182407408952713,0.9994791895151138,0.9990740716457367],"occluded":false},{"timestamp":13.880533333333334,"box":[0.8302083611488342,0.1509259194135666,0.999479204416275,0.9990740269422531],"occluded":false},{"timestamp":14.147466666666668,"box":[0.8203125596046448,0.1240740716457367,0.9994792342185974,0.9990740716457367],"occluded":false},{"timestamp":14.4144,"box":[0.8151041865348816,0.13055555522441864,0.9994791895151138,0.9990740865468979],"occluded":false},{"timestamp":14.681333333333335,"box":[0.7979167103767395,0.14629629254341125,0.9994792193174362,0.9990740716457367],"occluded":false},{"timestamp":14.948266666666667,"box":[0.7901042103767395,0.1398148089647293,0.9994792193174362,0.9990740567445755],"occluded":false},{"timestamp":15.215200000000001,"box":[0.7942708730697632,0.13611111044883728,0.9994792193174362,0.9990740716457367],"occluded":false},{"timestamp":15.482133333333334,"box":[0.7859375476837158,0.13240741193294525,0.9994792193174362,0.9990740865468979],"occluded":false},{"timestamp":15.749066666666668,"box":[0.7718750238418579,0.1259259283542633,0.999479204416275,0.9990740716457367],"occluded":false},{"timestamp":16.016000000000002,"box":[0.7614583969116211,0.12314814329147339,0.9994792491197586,0.9990740418434143],"occluded":false},{"timestamp":16.282933333333332,"box":[0.7473958730697632,0.10185185074806213,0.9994792342185974,0.9990740716457367],"occluded":false},{"timestamp":16.549866666666667,"box":[0.7234375476837158,0.10740740597248077,0.9994792342185974,0.9990740567445755],"occluded":false}],"mean_score":0.96875},{"noun_phrase":"man","object_id":2,"presentation_id":3,"observations":[{"timestamp":0.0,"box":[0.3281250298023224,0.24537037312984467,0.5427083671092987,0.9907407015562057],"occluded":false},{"timestamp":0.26693333333333336,"box":[0.33125001192092896,0.22962962090969086,0.5505208522081375,0.9990740269422531],"occluded":false},{"timestamp":0.5338666666666667,"box":[0.32864585518836975,0.26851850748062134,0.5328125357627869,0.9990740418434143],"occluded":false},{"timestamp":0.8008000000000001,"box":[0.34322917461395264,0.2787036895751953,0.5515625178813934,0.9990740418434143],"occluded":false},{"timestamp":1.0677333333333334,"box":[0.35468751192092896,0.2787036895751953,0.5661458522081375,0.9990740418434143],"occluded":false},{"timestamp":1.3346666666666667,"box":[0.3604166805744171,0.25462961196899414,0.5713541954755783,0.9990740418434143],"occluded":false},{"timestamp":1.6016000000000001,"box":[0.3854166865348816,0.27129629254341125,0.5937500298023224,0.9990740716457367],"occluded":false},{"timestamp":1.8685333333333334,"box":[0.39791667461395264,0.287962943315506,0.6098958551883698,0.9990740120410919],"occluded":false},{"timestamp":2.135466666666667,"box":[0.4088541865348816,0.2898148000240326,0.622395858168602,0.9990740716457367],"occluded":false},{"timestamp":2.4024,"box":[0.3973958492279053,0.2916666567325592,0.5973958671092987,0.9990740716457367],"occluded":false},{"timestamp":2.6693333333333333,"box":[0.45104169845581055,0.2611111104488373,0.6598958820104599,0.9990740716457367],"occluded":false},{"timestamp":2.936266666666667,"box":[0.4739583432674408,0.24351851642131805,0.6781250238418579,0.9990740865468979],"occluded":false},{"timestamp":3.2032000000000003,"box":[0.4895833730697632,0.2351851761341095,0.6921875476837158,0.9990740716457367],"occluded":false},{"timestamp":3.4701333333333335,"box":[0.4989583492279053,0.2620370388031006,0.6994791924953461,0.9990740418434143],"occluded":false},{"timestamp":3.7370666666666668,"box":[0.5036458373069763,0.28611111640930176,0.7093750089406967,0.9990740418434143],"occluded":false},{"timestamp":4.0040000000000004,"box":[0.5026041865348816,0.2916666567325592,0.7130208611488342,0.9990740716457367],"occluded":false},{"timestamp":4.270933333333334,"box":[0.4932291805744171,0.2898148000240326,0.6984375268220901,0.9972222149372101],"occluded":false},{"timestamp":4.537866666666667,"box":[0.49375003576278687,0.2944444417953491,0.7041667103767395,0.9990740418434143],"occluded":false},{"timestamp":4.8048,"box":[0.46250003576278687,0.2953703701496124,0.6755208820104599,0.9990740716457367],"occluded":false},{"timestamp":5.071733333333333,"box":[0.44062501192092896,0.2657407224178314,0.669270858168602,0.9990740716457367],"occluded":false},{"timestamp":5.338666666666667,"box":[0.42500001192092896,0.23240740597248077,0.6760416924953461,0.9990740567445755],"occluded":false},{"timestamp":5.6056,"box":[0.3984375298023224,0.21296295523643494,0.658854216337204,0.9990740716457367],"occluded":false},{"timestamp":5.872533333333334,"box":[0.38489586114883423,0.18888889253139496,0.6484375298023224,0.9990740865468979],"occluded":false},{"timestamp":6.139466666666667,"box":[0.3671875298023224,0.1814814805984497,0.6541666984558105,0.9990740418434143],"occluded":false},{"timestamp":6.4064000000000005,"box":[0.3411458432674408,0.1657407432794571,0.632291704416275,0.9990740567445755],"occluded":false},{"timestamp":6.673333333333334,"box":[0.3973958492279053,0.21759258210659027,0.6567708551883698,0.9990740269422531],"occluded":false},{"timestamp":6.940266666666667,"box":[0.42239585518836975,0.27037036418914795,0.6718750298023224,0.9990740418434143],"occluded":false},{"timestamp":7.2072,"box":[0.43906253576278687,0.31111109256744385,0.6911458969116211,0.9990740418434143],"occluded":false},{"timestamp":7.4741333333333335,"box":[0.460416704416275,0.3629629611968994,0.6968750506639481,0.9990740418434143],"occluded":false},{"timestamp":7.741066666666667,"box":[0.4635416865348816,0.41018518805503845,0.695833370089531,0.9990740716457367],"occluded":false},{"timestamp":8.008000000000001,"box":[0.45729169249534607,0.4037036895751953,0.688020870089531,0.9990740418434143],"occluded":false},{"timestamp":8.274933333333333,"box":[0.46302086114883423,0.3712962865829468,0.6921875327825546,0.9990740418434143],"occluded":false},{"timestamp":8.541866666666667,"box":[0.4437500238418579,0.37870368361473083,0.6656250357627869,0.9870370328426361],"occluded":false},{"timestamp":8.8088,"box":[0.4515625238418579,0.38240739703178406,0.6890625357627869,0.9861110746860504],"occluded":false},{"timestamp":9.075733333333334,"box":[0.44843751192092896,0.3444444239139557,0.6812500208616257,0.9990740716457367],"occluded":false},{"timestamp":9.342666666666666,"box":[0.46927085518836975,0.3296296298503876,0.7229166924953461,0.9990740716457367],"occluded":false},{"timestamp":9.6096,"box":[0.46562501788139343,0.3185185194015503,0.7239583730697632,0.9990740418434143],"occluded":false},{"timestamp":9.876533333333334,"box":[0.4531250298023224,0.33703702688217163,0.7156250476837158,0.9990740418434143],"occluded":false},{"timestamp":10.143466666666667,"box":[0.43020835518836975,0.34074074029922485,0.6864583790302277,0.9990740418434143],"occluded":false},{"timestamp":10.410400000000001,"box":[0.41041669249534607,0.3212963044643402,0.6682291924953461,0.9990740716457367],"occluded":false},{"timestamp":10.677333333333333,"box":[0.4348958432674408,0.3046296238899231,0.6958333551883698,0.9990740418434143],"occluded":false},{"timestamp":10.944266666666667,"box":[0.44583335518836975,0.3027777671813965,0.7046875357627869,0.9990740418434143],"occluded":false},{"timestamp":11.2112,"box":[0.4750000238418579,0.3212963044643402,0.7276042103767395,0.9990740716457367],"occluded":false},{"timestamp":11.478133333333334,"box":[0.5078125,0.3314814865589142,0.765625,0.9990740716457367],"occluded":false},{"timestamp":11.745066666666668,"box":[0.5385416746139526,0.34351852536201477,0.7953125238418579,0.9990740716457367],"occluded":false},{"timestamp":12.012,"box":[0.5348958373069763,0.32592591643333435,0.792187511920929,0.9990740716457367],"occluded":false},{"timestamp":12.278933333333335,"box":[0.510937511920929,0.3083333373069763,0.7661458551883698,0.9990740418434143],"occluded":false},{"timestamp":12.545866666666667,"box":[0.5057291984558105,0.2944444417953491,0.7588542103767395,0.9990740418434143],"occluded":false},{"timestamp":12.812800000000001,"box":[0.4984375238418579,0.29814815521240234,0.7468750327825546,0.9972222447395325],"occluded":false},{"timestamp":13.079733333333333,"box":[0.49635419249534607,0.3055555522441864,0.7447917014360428,0.9990740716457367],"occluded":false},{"timestamp":13.346666666666668,"box":[0.491666704416275,0.29814815521240234,0.7375000566244125,0.9990740418434143],"occluded":false},{"timestamp":13.6136,"box":[0.49270835518836975,0.3046296238899231,0.741666704416275,0.9990740418434143],"occluded":false},{"timestamp":13.880533333333334,"box":[0.4869791865348816,0.29722222685813904,0.7385416924953461,0.9990740716457367],"occluded":false},{"timestamp":14.147466666666668,"box":[0.48229169845581055,0.2898148000240326,0.7343750596046448,0.9990740716457367],"occluded":false},{"timestamp":14.4144,"box":[0.4776041805744171,0.3046296238899231,0.7244791984558105,0.9990740418434143],"occluded":false},{"timestamp":14.681333333333335,"box":[0.46562501788139343,0.31203702092170715,0.7020833641290665,0.9990740716457367],"occluded":false},{"timestamp":14.948266666666667,"box":[0.4645833671092987,0.3064814805984497,0.7036458849906921,0.9990740418434143],"occluded":false},{"timestamp":15.215200000000001,"box":[0.46250003576278687,0.3166666626930237,0.7052083760499954,0.9990740418434143],"occluded":false},{"timestamp":15.482133333333334,"box":[0.4531250298023224,0.31296294927597046,0.6989583820104599,0.9990740418434143],"occluded":false},{"timestamp":15.749066666666668,"box":[0.44322919845581055,0.31018519401550293,0.6911458820104599,0.9990740418434143],"occluded":false},{"timestamp":16.016000000000002,"box":[0.4322916865348816,0.31388887763023376,0.6812500357627869,0.9990740716457367],"occluded":false},{"timestamp":16.282933333333332,"box":[0.41979169845581055,0.31111109256744385,0.6703125536441803,0.9990740418434143],"occluded":false},{"timestamp":16.549866666666667,"box":[0.3958333432674408,0.3157407343387604,0.6447916924953461,0.9990740716457367],"occluded":false}],"mean_score":0.941605806350708}],"dropped":{},"fates":[{"noun_phrase":"man","object_id":0,"frames":63,"mean_score":0.96875,"mean_area":0.199678483459795,"outcome":"kept","presentation_id":2,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"man","object_id":1,"frames":63,"mean_score":0.9609756469726562,"mean_area":0.5862247205170363,"outcome":"kept","presentation_id":1,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"man","object_id":2,"frames":63,"mean_score":0.941605806350708,"mean_area":0.1670317607746074,"outcome":"kept","presentation_id":3,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null}],"gates":{"min_track_frames":3,"min_mean_score":0.5,"merge_iou":0.8,"max_tracklets":12}}},"render":{"inputs":{"tracklets":3,"as_video":true,"fps":1.0,"max_frames":30},"outputs":{"media":"/home/zzk/data/video_bbox/vlm_gemini3.5_lite/renders/--X6fL8DOMM_chunk_000_scene_003_choices.mp4","media_bytes":5775150,"frames":[{"index":1,"timestamp":0.0,"file":"--X6fL8DOMM_chunk_000_scene_003_choices.mp4"},{"index":2,"timestamp":1.0677333333333334,"file":"--X6fL8DOMM_chunk_000_scene_003_choices.mp4"},{"index":3,"timestamp":1.8685333333333334,"file":"--X6fL8DOMM_chunk_000_scene_003_choices.mp4"},{"index":4,"timestamp":2.936266666666667,"file":"--X6fL8DOMM_chunk_000_scene_003_choices.mp4"},{"index":5,"timestamp":4.0040000000000004,"file":"--X6fL8DOMM_chunk_000_scene_003_choices.mp4"},{"index":6,"timestamp":4.8048,"file":"--X6fL8DOMM_chunk_000_scene_003_choices.mp4"},{"index":7,"timestamp":5.872533333333334,"file":"--X6fL8DOMM_chunk_000_scene_003_choices.mp4"},{"index":8,"timestamp":6.940266666666667,"file":"--X6fL8DOMM_chunk_000_scene_003_choices.mp4"},{"index":9,"timestamp":7.741066666666667,"file":"--X6fL8DOMM_chunk_000_scene_003_choices.mp4"},{"index":10,"timestamp":8.8088,"file":"--X6fL8DOMM_chunk_000_scene_003_choices.mp4"},{"index":11,"timestamp":9.6096,"file":"--X6fL8DOMM_chunk_000_scene_003_choices.mp4"},{"index":12,"timestamp":10.677333333333333,"file":"--X6fL8DOMM_chunk_000_scene_003_choices.mp4"},{"index":13,"timestamp":11.745066666666668,"file":"--X6fL8DOMM_chunk_000_scene_003_choices.mp4"},{"index":14,"timestamp":12.545866666666667,"file":"--X6fL8DOMM_chunk_000_scene_003_choices.mp4"},{"index":15,"timestamp":13.6136,"file":"--X6fL8DOMM_chunk_000_scene_003_choices.mp4"},{"index":16,"timestamp":14.681333333333335,"file":"--X6fL8DOMM_chunk_000_scene_003_choices.mp4"},{"index":17,"timestamp":15.482133333333334,"file":"--X6fL8DOMM_chunk_000_scene_003_choices.mp4"},{"index":18,"timestamp":16.549866666666667,"file":"--X6fL8DOMM_chunk_000_scene_003_choices.mp4"}]}},"select":{"inputs":{"model":"gemini-3.5-flash-lite","temperature":0.0,"as_video":true,"media_bytes":5775150,"media":["--X6fL8DOMM_chunk_000_scene_003_choices.mp4"],"choice_frames":18,"system_prompt":"You are labelling video data for referring multi-object tracking.\n\nYou will see a numbered sequence of frames from one clip, in time order. Each frame has its own number and timestamp printed in the top-left corner. Candidate objects have been detected for you and drawn on the frames as coloured boxes, each labelled with a number. The same object keeps the same number and the same colour on every frame. Below each frame is a strip of enlarged crops of those same numbered boxes, so you can identify a small or distant object.\n\nFor each description you are given, answer two questions:\n\n1. MATCHING_IDS — which of the numbered objects the description refers to. List every number that matches, or an EMPTY LIST if none of them do. An empty list is a correct and expected answer: some descriptions are deliberately about something that is not in this clip. Do not pick a number just to have picked one, and do not pick a number for an object that merely resembles the description.\n\n2. START_FRAME and END_FRAME — the frame numbers between which the described state actually holds. If the description is \"the car that's turning left\", the interval is the frames in which that car is turning, NOT every frame in which the car is visible. Use the frame numbers printed on the frames. START_FRAME must not be greater than END_FRAME. When MATCHING_IDS is empty, give the full range (the first and last frame numbers); it is ignored.\n\n3. CONFIDENCE — how sure you are of the MATCHING_IDS answer, from 0.0 to 1.0. Be honest: a low number costs the example nothing, and an overconfident wrong answer becomes a wrong label.\n\nRules:\n- Only use numbers that are actually drawn on the frames.\n- Never output coordinates, box values, or pixel positions.\n- Judge motion by comparing the object's position across frames, not by guessing from one frame.\n","user_prompt":"The video has 18 frames, numbered 1 to 18. Each frame is one moment of the clip and has its number burned into the top left corner.\n\nAnswer for each of the following descriptions, in the order given:\n1. the man with sunglasses sitting in the front passenger seat\n2. the young man sitting in the back seat wearing a white t-shirt","questions":["the man with sunglasses sitting in the front passenger seat","the young man sitting in the back seat wearing a white t-shirt"]},"outputs":{"raw_response":"{\n  \"answers\": [\n    {\n      \"description_index\": 1,\n      \"matching_ids\": [\n        2\n      ],\n      \"start_frame\": 1,\n      \"end_frame\": 18,\n      \"confidence\": 1.0\n    },\n    {\n      \"description_index\": 2,\n      \"matching_ids\": [\n        3\n      ],\n      \"start_frame\": 1,\n      \"end_frame\": 18,\n      \"confidence\": 1.0\n    }\n  ]\n}","raw_answers":[{"description_index":1,"matching_ids":[2],"start_frame":1,"end_frame":18,"confidence":1.0},{"description_index":2,"matching_ids":[3],"start_frame":1,"end_frame":18,"confidence":1.0}],"answers":[{"expression":"the man with sunglasses sitting in the front passenger seat","matching_presentation_ids":[2],"start_choice_frame":1,"end_choice_frame":18,"confidence":1.0,"self_reported_confidence":1.0},{"expression":"the young man sitting in the back seat wearing a white t-shirt","matching_presentation_ids":[3],"start_choice_frame":1,"end_choice_frame":18,"confidence":1.0,"self_reported_confidence":1.0}]}},"verify":{"inputs":{"expressions":2,"tracklets":3,"selections":2,"gates":"interval, count, unknown ids, confidence, then geometry (motion)"},"outputs":{"kept":[{"expression":"the man with sunglasses sitting in the front passenger seat","kind":"spatial","is_negative":false,"tracks":[{"timestamps":[0.0,0.26693333333333336,0.5338666666666667,0.8008000000000001,1.0677333333333334,1.3346666666666667,1.6016000000000001,1.8685333333333334,2.135466666666667,2.4024,2.6693333333333333,2.936266666666667,3.2032000000000003,3.4701333333333335,3.7370666666666668,4.0040000000000004,4.270933333333334,4.537866666666667,4.8048,5.071733333333333,5.338666666666667,5.6056,5.872533333333334,6.139466666666667,6.4064000000000005,6.673333333333334,6.940266666666667,7.2072,7.4741333333333335,7.741066666666667,8.008000000000001,8.274933333333333,8.541866666666667,8.8088,9.075733333333334,9.342666666666666,9.6096,9.876533333333334,10.143466666666667,10.410400000000001,10.677333333333333,10.944266666666667,11.2112,11.478133333333334,11.745066666666668,12.012,12.278933333333335,12.545866666666667,12.812800000000001,13.079733333333333,13.346666666666668,13.6136,13.880533333333334,14.147466666666668,14.4144,14.681333333333335,14.948266666666667,15.215200000000001,15.482133333333334,15.749066666666668,16.016000000000002,16.282933333333332,16.549866666666667],"boxes":[[0.6380208730697632,0.10648147761821747,0.9994792342185974,0.9990740269422531],[0.6291667222976685,0.11203703284263611,0.9994792342185974,0.9990740716457367],[0.6177083849906921,0.15648147463798523,0.9994792342185974,0.9990740716457367],[0.6411458849906921,0.1518518477678299,0.9994792342185974,0.9990740567445755],[0.6598958969116211,0.1666666567325592,0.9994792342185974,0.9990740716457367],[0.518750011920929,0.14629629254341125,0.999479204416275,0.9990740716457367],[0.7005208730697632,0.16944444179534912,0.9994792342185974,0.9990740418434143],[0.7260417342185974,0.1796296238899231,0.9994792342185974,0.9990740418434143],[0.7494792342185974,0.17685185372829437,0.9994792342185974,0.9990740865468979],[0.5604166984558105,0.1648148149251938,0.9994792342185974,0.9990740865468979],[0.7864583730697632,0.1379629671573639,0.9994792193174362,0.9990740716457367],[0.8015625476837158,0.125,0.9994792193174362,0.9990740418434143],[0.8213542103767395,0.5370370149612427,0.9994792193174362,0.9990740418434143],[0.8291667103767395,0.555555522441864,0.9994792193174362,0.9990740418434143],[0.8447917103767395,0.5916666388511658,0.9994792193174362,0.9990740418434143],[0.8437500596046448,0.5990740656852722,0.9994792342185974,0.9990740716457367],[0.8291667103767395,0.2518518567085266,0.9994792193174362,0.9990740418434143],[0.8338541984558105,0.2666666507720947,0.999479204416275,0.9990740418434143],[0.7984375357627869,0.17499999701976776,0.9994792193174362,0.9990740269422531],[0.7880208492279053,0.1379629671573639,0.9994791895151138,0.9990740716457367],[0.7890625596046448,0.07592592388391495,0.9994792342185974,0.9990740790963173],[0.7671875357627869,0.05000000074505806,0.9994792193174362,0.9990740306675434],[0.7484375238418579,0.01666666567325592,0.999479204416275,0.9990740567445755],[0.7541667222976685,0.0,0.9994792342185974,0.9990740418434143],[0.7385417222976685,0.0,0.9994792342185974,0.9990740418434143],[0.7755208611488342,0.0,0.999479204416275,0.9990740418434143],[0.7984375357627869,0.06111111119389534,0.9994792193174362,0.9990740604698658],[0.8281250596046448,0.3314814865589142,0.9994792342185974,0.9990740716457367],[0.8526042103767395,0.6444444060325623,0.9994792193174362,0.9990740418434143],[0.8708333969116211,0.7074074149131775,0.9994792342185974,0.9990740716457367],[0.8593750596046448,0.6814814805984497,0.9994792342185974,0.9990740716457367],[0.8473958969116211,0.6037036776542664,0.9994792342185974,0.9990740418434143],[0.8177083730697632,0.5222222208976746,0.9994792193174362,0.9990740716457367],[0.8119792342185974,0.18703703582286835,0.9994792491197586,0.9990740865468979],[0.7984375357627869,0.05092592537403107,0.9994792193174362,0.9990740567445755],[0.8109375238418579,0.031481482088565826,0.999479204416275,0.9990740790963173],[0.5328125357627869,0.010185184888541698,0.9994792342185974,0.999074044637382],[0.4911458492279053,0.03888888657093048,0.9994792342185974,0.9990740567445755],[0.5328125357627869,0.07129629701375961,0.9994792342185974,0.9990740641951561],[0.7354167103767395,0.040740739554166794,0.9994792342185974,0.9990740530192852],[0.4911458492279053,0.014814814552664757,0.9994792342185974,0.999074062332511],[0.768750011920929,0.028703702613711357,0.9994791895151138,0.9990740548819304],[0.5744792222976685,0.07222221791744232,0.9994792342185974,0.9990740269422531],[0.8463541865348816,0.1259259283542633,0.9994791895151138,0.9990740716457367],[0.8848958611488342,0.6361110806465149,0.9994791969656944,0.9990740418434143],[0.8802083730697632,0.6305555701255798,0.9994792118668556,0.9990740716457367],[0.8526042103767395,0.17407406866550446,0.9994792193174362,0.9990740567445755],[0.8442708849906921,0.14722222089767456,0.9994792193174362,0.9990740418434143],[0.8354167342185974,0.1518518477678299,0.9994792491197586,0.9990740567445755],[0.8375000357627869,0.1675925850868225,0.999479204416275,0.9990740418434143],[0.8276042342185974,0.17037037014961243,0.9994792491197586,0.9990740716457367],[0.8348958492279053,0.182407408952713,0.9994791895151138,0.9990740716457367],[0.8302083611488342,0.1509259194135666,0.999479204416275,0.9990740269422531],[0.8203125596046448,0.1240740716457367,0.9994792342185974,0.9990740716457367],[0.8151041865348816,0.13055555522441864,0.9994791895151138,0.9990740865468979],[0.7979167103767395,0.14629629254341125,0.9994792193174362,0.9990740716457367],[0.7901042103767395,0.1398148089647293,0.9994792193174362,0.9990740567445755],[0.7942708730697632,0.13611111044883728,0.9994792193174362,0.9990740716457367],[0.7859375476837158,0.13240741193294525,0.9994792193174362,0.9990740865468979],[0.7718750238418579,0.1259259283542633,0.999479204416275,0.9990740716457367],[0.7614583969116211,0.12314814329147339,0.9994792491197586,0.9990740418434143],[0.7473958730697632,0.10185185074806213,0.9994792342185974,0.9990740716457367],[0.7234375476837158,0.10740740597248077,0.9994792342185974,0.9990740567445755]]}]},{"expression":"the young man sitting in the back seat wearing a white t-shirt","kind":"spatial","is_negative":false,"tracks":[{"timestamps":[0.0,0.26693333333333336,0.5338666666666667,0.8008000000000001,1.0677333333333334,1.3346666666666667,1.6016000000000001,1.8685333333333334,2.135466666666667,2.4024,2.6693333333333333,2.936266666666667,3.2032000000000003,3.4701333333333335,3.7370666666666668,4.0040000000000004,4.270933333333334,4.537866666666667,4.8048,5.071733333333333,5.338666666666667,5.6056,5.872533333333334,6.139466666666667,6.4064000000000005,6.673333333333334,6.940266666666667,7.2072,7.4741333333333335,7.741066666666667,8.008000000000001,8.274933333333333,8.541866666666667,8.8088,9.075733333333334,9.342666666666666,9.6096,9.876533333333334,10.143466666666667,10.410400000000001,10.677333333333333,10.944266666666667,11.2112,11.478133333333334,11.745066666666668,12.012,12.278933333333335,12.545866666666667,12.812800000000001,13.079733333333333,13.346666666666668,13.6136,13.880533333333334,14.147466666666668,14.4144,14.681333333333335,14.948266666666667,15.215200000000001,15.482133333333334,15.749066666666668,16.016000000000002,16.282933333333332,16.549866666666667],"boxes":[[0.3281250298023224,0.24537037312984467,0.5427083671092987,0.9907407015562057],[0.33125001192092896,0.22962962090969086,0.5505208522081375,0.9990740269422531],[0.32864585518836975,0.26851850748062134,0.5328125357627869,0.9990740418434143],[0.34322917461395264,0.2787036895751953,0.5515625178813934,0.9990740418434143],[0.35468751192092896,0.2787036895751953,0.5661458522081375,0.9990740418434143],[0.3604166805744171,0.25462961196899414,0.5713541954755783,0.9990740418434143],[0.3854166865348816,0.27129629254341125,0.5937500298023224,0.9990740716457367],[0.39791667461395264,0.287962943315506,0.6098958551883698,0.9990740120410919],[0.4088541865348816,0.2898148000240326,0.622395858168602,0.9990740716457367],[0.3973958492279053,0.2916666567325592,0.5973958671092987,0.9990740716457367],[0.45104169845581055,0.2611111104488373,0.6598958820104599,0.9990740716457367],[0.4739583432674408,0.24351851642131805,0.6781250238418579,0.9990740865468979],[0.4895833730697632,0.2351851761341095,0.6921875476837158,0.9990740716457367],[0.4989583492279053,0.2620370388031006,0.6994791924953461,0.9990740418434143],[0.5036458373069763,0.28611111640930176,0.7093750089406967,0.9990740418434143],[0.5026041865348816,0.2916666567325592,0.7130208611488342,0.9990740716457367],[0.4932291805744171,0.2898148000240326,0.6984375268220901,0.9972222149372101],[0.49375003576278687,0.2944444417953491,0.7041667103767395,0.9990740418434143],[0.46250003576278687,0.2953703701496124,0.6755208820104599,0.9990740716457367],[0.44062501192092896,0.2657407224178314,0.669270858168602,0.9990740716457367],[0.42500001192092896,0.23240740597248077,0.6760416924953461,0.9990740567445755],[0.3984375298023224,0.21296295523643494,0.658854216337204,0.9990740716457367],[0.38489586114883423,0.18888889253139496,0.6484375298023224,0.9990740865468979],[0.3671875298023224,0.1814814805984497,0.6541666984558105,0.9990740418434143],[0.3411458432674408,0.1657407432794571,0.632291704416275,0.9990740567445755],[0.3973958492279053,0.21759258210659027,0.6567708551883698,0.9990740269422531],[0.42239585518836975,0.27037036418914795,0.6718750298023224,0.9990740418434143],[0.43906253576278687,0.31111109256744385,0.6911458969116211,0.9990740418434143],[0.460416704416275,0.3629629611968994,0.6968750506639481,0.9990740418434143],[0.4635416865348816,0.41018518805503845,0.695833370089531,0.9990740716457367],[0.45729169249534607,0.4037036895751953,0.688020870089531,0.9990740418434143],[0.46302086114883423,0.3712962865829468,0.6921875327825546,0.9990740418434143],[0.4437500238418579,0.37870368361473083,0.6656250357627869,0.9870370328426361],[0.4515625238418579,0.38240739703178406,0.6890625357627869,0.9861110746860504],[0.44843751192092896,0.3444444239139557,0.6812500208616257,0.9990740716457367],[0.46927085518836975,0.3296296298503876,0.7229166924953461,0.9990740716457367],[0.46562501788139343,0.3185185194015503,0.7239583730697632,0.9990740418434143],[0.4531250298023224,0.33703702688217163,0.7156250476837158,0.9990740418434143],[0.43020835518836975,0.34074074029922485,0.6864583790302277,0.9990740418434143],[0.41041669249534607,0.3212963044643402,0.6682291924953461,0.9990740716457367],[0.4348958432674408,0.3046296238899231,0.6958333551883698,0.9990740418434143],[0.44583335518836975,0.3027777671813965,0.7046875357627869,0.9990740418434143],[0.4750000238418579,0.3212963044643402,0.7276042103767395,0.9990740716457367],[0.5078125,0.3314814865589142,0.765625,0.9990740716457367],[0.5385416746139526,0.34351852536201477,0.7953125238418579,0.9990740716457367],[0.5348958373069763,0.32592591643333435,0.792187511920929,0.9990740716457367],[0.510937511920929,0.3083333373069763,0.7661458551883698,0.9990740418434143],[0.5057291984558105,0.2944444417953491,0.7588542103767395,0.9990740418434143],[0.4984375238418579,0.29814815521240234,0.7468750327825546,0.9972222447395325],[0.49635419249534607,0.3055555522441864,0.7447917014360428,0.9990740716457367],[0.491666704416275,0.29814815521240234,0.7375000566244125,0.9990740418434143],[0.49270835518836975,0.3046296238899231,0.741666704416275,0.9990740418434143],[0.4869791865348816,0.29722222685813904,0.7385416924953461,0.9990740716457367],[0.48229169845581055,0.2898148000240326,0.7343750596046448,0.9990740716457367],[0.4776041805744171,0.3046296238899231,0.7244791984558105,0.9990740418434143],[0.46562501788139343,0.31203702092170715,0.7020833641290665,0.9990740716457367],[0.4645833671092987,0.3064814805984497,0.7036458849906921,0.9990740418434143],[0.46250003576278687,0.3166666626930237,0.7052083760499954,0.9990740418434143],[0.4531250298023224,0.31296294927597046,0.6989583820104599,0.9990740418434143],[0.44322919845581055,0.31018519401550293,0.6911458820104599,0.9990740418434143],[0.4322916865348816,0.31388887763023376,0.6812500357627869,0.9990740716457367],[0.41979169845581055,0.31111109256744385,0.6703125536441803,0.9990740418434143],[0.3958333432674408,0.3157407343387604,0.6447916924953461,0.9990740716457367]]}]}],"rejected":{},"rejection_counts":{},"row":{"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--X6fL8DOMM_chunk_000_scene_003.mp4","scene_key":"--X6fL8DOMM_chunk_000_scene_003","category":"test_vehicles","height":1080,"width":1920,"fps":29.97002997002997,"num_frames":503,"duration":16.783433333333335,"rmot_expressions":[{"expression":"the man with sunglasses sitting in the front passenger seat","kind":"spatial","is_negative":false,"tracks":[{"timestamps":[0.0,0.26693333333333336,0.5338666666666667,0.8008000000000001,1.0677333333333334,1.3346666666666667,1.6016000000000001,1.8685333333333334,2.135466666666667,2.4024,2.6693333333333333,2.936266666666667,3.2032000000000003,3.4701333333333335,3.7370666666666668,4.0040000000000004,4.270933333333334,4.537866666666667,4.8048,5.071733333333333,5.338666666666667,5.6056,5.872533333333334,6.139466666666667,6.4064000000000005,6.673333333333334,6.940266666666667,7.2072,7.4741333333333335,7.741066666666667,8.008000000000001,8.274933333333333,8.541866666666667,8.8088,9.075733333333334,9.342666666666666,9.6096,9.876533333333334,10.143466666666667,10.410400000000001,10.677333333333333,10.944266666666667,11.2112,11.478133333333334,11.745066666666668,12.012,12.278933333333335,12.545866666666667,12.812800000000001,13.079733333333333,13.346666666666668,13.6136,13.880533333333334,14.147466666666668,14.4144,14.681333333333335,14.948266666666667,15.215200000000001,15.482133333333334,15.749066666666668,16.016000000000002,16.282933333333332,16.549866666666667],"boxes":[[0.6380208730697632,0.10648147761821747,0.9994792342185974,0.9990740269422531],[0.6291667222976685,0.11203703284263611,0.9994792342185974,0.9990740716457367],[0.6177083849906921,0.15648147463798523,0.9994792342185974,0.9990740716457367],[0.6411458849906921,0.1518518477678299,0.9994792342185974,0.9990740567445755],[0.6598958969116211,0.1666666567325592,0.9994792342185974,0.9990740716457367],[0.518750011920929,0.14629629254341125,0.999479204416275,0.9990740716457367],[0.7005208730697632,0.16944444179534912,0.9994792342185974,0.9990740418434143],[0.7260417342185974,0.1796296238899231,0.9994792342185974,0.9990740418434143],[0.7494792342185974,0.17685185372829437,0.9994792342185974,0.9990740865468979],[0.5604166984558105,0.1648148149251938,0.9994792342185974,0.9990740865468979],[0.7864583730697632,0.1379629671573639,0.9994792193174362,0.9990740716457367],[0.8015625476837158,0.125,0.9994792193174362,0.9990740418434143],[0.8213542103767395,0.5370370149612427,0.9994792193174362,0.9990740418434143],[0.8291667103767395,0.555555522441864,0.9994792193174362,0.9990740418434143],[0.8447917103767395,0.5916666388511658,0.9994792193174362,0.9990740418434143],[0.8437500596046448,0.5990740656852722,0.9994792342185974,0.9990740716457367],[0.8291667103767395,0.2518518567085266,0.9994792193174362,0.9990740418434143],[0.8338541984558105,0.2666666507720947,0.999479204416275,0.9990740418434143],[0.7984375357627869,0.17499999701976776,0.9994792193174362,0.9990740269422531],[0.7880208492279053,0.1379629671573639,0.9994791895151138,0.9990740716457367],[0.7890625596046448,0.07592592388391495,0.9994792342185974,0.9990740790963173],[0.7671875357627869,0.05000000074505806,0.9994792193174362,0.9990740306675434],[0.7484375238418579,0.01666666567325592,0.999479204416275,0.9990740567445755],[0.7541667222976685,0.0,0.9994792342185974,0.9990740418434143],[0.7385417222976685,0.0,0.9994792342185974,0.9990740418434143],[0.7755208611488342,0.0,0.999479204416275,0.9990740418434143],[0.7984375357627869,0.06111111119389534,0.9994792193174362,0.9990740604698658],[0.8281250596046448,0.3314814865589142,0.9994792342185974,0.9990740716457367],[0.8526042103767395,0.6444444060325623,0.9994792193174362,0.9990740418434143],[0.8708333969116211,0.7074074149131775,0.9994792342185974,0.9990740716457367],[0.8593750596046448,0.6814814805984497,0.9994792342185974,0.9990740716457367],[0.8473958969116211,0.6037036776542664,0.9994792342185974,0.9990740418434143],[0.8177083730697632,0.5222222208976746,0.9994792193174362,0.9990740716457367],[0.8119792342185974,0.18703703582286835,0.9994792491197586,0.9990740865468979],[0.7984375357627869,0.05092592537403107,0.9994792193174362,0.9990740567445755],[0.8109375238418579,0.031481482088565826,0.999479204416275,0.9990740790963173],[0.5328125357627869,0.010185184888541698,0.9994792342185974,0.999074044637382],[0.4911458492279053,0.03888888657093048,0.9994792342185974,0.9990740567445755],[0.5328125357627869,0.07129629701375961,0.9994792342185974,0.9990740641951561],[0.7354167103767395,0.040740739554166794,0.9994792342185974,0.9990740530192852],[0.4911458492279053,0.014814814552664757,0.9994792342185974,0.999074062332511],[0.768750011920929,0.028703702613711357,0.9994791895151138,0.9990740548819304],[0.5744792222976685,0.07222221791744232,0.9994792342185974,0.9990740269422531],[0.8463541865348816,0.1259259283542633,0.9994791895151138,0.9990740716457367],[0.8848958611488342,0.6361110806465149,0.9994791969656944,0.9990740418434143],[0.8802083730697632,0.6305555701255798,0.9994792118668556,0.9990740716457367],[0.8526042103767395,0.17407406866550446,0.9994792193174362,0.9990740567445755],[0.8442708849906921,0.14722222089767456,0.9994792193174362,0.9990740418434143],[0.8354167342185974,0.1518518477678299,0.9994792491197586,0.9990740567445755],[0.8375000357627869,0.1675925850868225,0.999479204416275,0.9990740418434143],[0.8276042342185974,0.17037037014961243,0.9994792491197586,0.9990740716457367],[0.8348958492279053,0.182407408952713,0.9994791895151138,0.9990740716457367],[0.8302083611488342,0.1509259194135666,0.999479204416275,0.9990740269422531],[0.8203125596046448,0.1240740716457367,0.9994792342185974,0.9990740716457367],[0.8151041865348816,0.13055555522441864,0.9994791895151138,0.9990740865468979],[0.7979167103767395,0.14629629254341125,0.9994792193174362,0.9990740716457367],[0.7901042103767395,0.1398148089647293,0.9994792193174362,0.9990740567445755],[0.7942708730697632,0.13611111044883728,0.9994792193174362,0.9990740716457367],[0.7859375476837158,0.13240741193294525,0.9994792193174362,0.9990740865468979],[0.7718750238418579,0.1259259283542633,0.999479204416275,0.9990740716457367],[0.7614583969116211,0.12314814329147339,0.9994792491197586,0.9990740418434143],[0.7473958730697632,0.10185185074806213,0.9994792342185974,0.9990740716457367],[0.7234375476837158,0.10740740597248077,0.9994792342185974,0.9990740567445755]]}]},{"expression":"the young man sitting in the back seat wearing a white t-shirt","kind":"spatial","is_negative":false,"tracks":[{"timestamps":[0.0,0.26693333333333336,0.5338666666666667,0.8008000000000001,1.0677333333333334,1.3346666666666667,1.6016000000000001,1.8685333333333334,2.135466666666667,2.4024,2.6693333333333333,2.936266666666667,3.2032000000000003,3.4701333333333335,3.7370666666666668,4.0040000000000004,4.270933333333334,4.537866666666667,4.8048,5.071733333333333,5.338666666666667,5.6056,5.872533333333334,6.139466666666667,6.4064000000000005,6.673333333333334,6.940266666666667,7.2072,7.4741333333333335,7.741066666666667,8.008000000000001,8.274933333333333,8.541866666666667,8.8088,9.075733333333334,9.342666666666666,9.6096,9.876533333333334,10.143466666666667,10.410400000000001,10.677333333333333,10.944266666666667,11.2112,11.478133333333334,11.745066666666668,12.012,12.278933333333335,12.545866666666667,12.812800000000001,13.079733333333333,13.346666666666668,13.6136,13.880533333333334,14.147466666666668,14.4144,14.681333333333335,14.948266666666667,15.215200000000001,15.482133333333334,15.749066666666668,16.016000000000002,16.282933333333332,16.549866666666667],"boxes":[[0.3281250298023224,0.24537037312984467,0.5427083671092987,0.9907407015562057],[0.33125001192092896,0.22962962090969086,0.5505208522081375,0.9990740269422531],[0.32864585518836975,0.26851850748062134,0.5328125357627869,0.9990740418434143],[0.34322917461395264,0.2787036895751953,0.5515625178813934,0.9990740418434143],[0.35468751192092896,0.2787036895751953,0.5661458522081375,0.9990740418434143],[0.3604166805744171,0.25462961196899414,0.5713541954755783,0.9990740418434143],[0.3854166865348816,0.27129629254341125,0.5937500298023224,0.9990740716457367],[0.39791667461395264,0.287962943315506,0.6098958551883698,0.9990740120410919],[0.4088541865348816,0.2898148000240326,0.622395858168602,0.9990740716457367],[0.3973958492279053,0.2916666567325592,0.5973958671092987,0.9990740716457367],[0.45104169845581055,0.2611111104488373,0.6598958820104599,0.9990740716457367],[0.4739583432674408,0.24351851642131805,0.6781250238418579,0.9990740865468979],[0.4895833730697632,0.2351851761341095,0.6921875476837158,0.9990740716457367],[0.4989583492279053,0.2620370388031006,0.6994791924953461,0.9990740418434143],[0.5036458373069763,0.28611111640930176,0.7093750089406967,0.9990740418434143],[0.5026041865348816,0.2916666567325592,0.7130208611488342,0.9990740716457367],[0.4932291805744171,0.2898148000240326,0.6984375268220901,0.9972222149372101],[0.49375003576278687,0.2944444417953491,0.7041667103767395,0.9990740418434143],[0.46250003576278687,0.2953703701496124,0.6755208820104599,0.9990740716457367],[0.44062501192092896,0.2657407224178314,0.669270858168602,0.9990740716457367],[0.42500001192092896,0.23240740597248077,0.6760416924953461,0.9990740567445755],[0.3984375298023224,0.21296295523643494,0.658854216337204,0.9990740716457367],[0.38489586114883423,0.18888889253139496,0.6484375298023224,0.9990740865468979],[0.3671875298023224,0.1814814805984497,0.6541666984558105,0.9990740418434143],[0.3411458432674408,0.1657407432794571,0.632291704416275,0.9990740567445755],[0.3973958492279053,0.21759258210659027,0.6567708551883698,0.9990740269422531],[0.42239585518836975,0.27037036418914795,0.6718750298023224,0.9990740418434143],[0.43906253576278687,0.31111109256744385,0.6911458969116211,0.9990740418434143],[0.460416704416275,0.3629629611968994,0.6968750506639481,0.9990740418434143],[0.4635416865348816,0.41018518805503845,0.695833370089531,0.9990740716457367],[0.45729169249534607,0.4037036895751953,0.688020870089531,0.9990740418434143],[0.46302086114883423,0.3712962865829468,0.6921875327825546,0.9990740418434143],[0.4437500238418579,0.37870368361473083,0.6656250357627869,0.9870370328426361],[0.4515625238418579,0.38240739703178406,0.6890625357627869,0.9861110746860504],[0.44843751192092896,0.3444444239139557,0.6812500208616257,0.9990740716457367],[0.46927085518836975,0.3296296298503876,0.7229166924953461,0.9990740716457367],[0.46562501788139343,0.3185185194015503,0.7239583730697632,0.9990740418434143],[0.4531250298023224,0.33703702688217163,0.7156250476837158,0.9990740418434143],[0.43020835518836975,0.34074074029922485,0.6864583790302277,0.9990740418434143],[0.41041669249534607,0.3212963044643402,0.6682291924953461,0.9990740716457367],[0.4348958432674408,0.3046296238899231,0.6958333551883698,0.9990740418434143],[0.44583335518836975,0.3027777671813965,0.7046875357627869,0.9990740418434143],[0.4750000238418579,0.3212963044643402,0.7276042103767395,0.9990740716457367],[0.5078125,0.3314814865589142,0.765625,0.9990740716457367],[0.5385416746139526,0.34351852536201477,0.7953125238418579,0.9990740716457367],[0.5348958373069763,0.32592591643333435,0.792187511920929,0.9990740716457367],[0.510937511920929,0.3083333373069763,0.7661458551883698,0.9990740418434143],[0.5057291984558105,0.2944444417953491,0.7588542103767395,0.9990740418434143],[0.4984375238418579,0.29814815521240234,0.7468750327825546,0.9972222447395325],[0.49635419249534607,0.3055555522441864,0.7447917014360428,0.9990740716457367],[0.491666704416275,0.29814815521240234,0.7375000566244125,0.9990740418434143],[0.49270835518836975,0.3046296238899231,0.741666704416275,0.9990740418434143],[0.4869791865348816,0.29722222685813904,0.7385416924953461,0.9990740716457367],[0.48229169845581055,0.2898148000240326,0.7343750596046448,0.9990740716457367],[0.4776041805744171,0.3046296238899231,0.7244791984558105,0.9990740418434143],[0.46562501788139343,0.31203702092170715,0.7020833641290665,0.9990740716457367],[0.4645833671092987,0.3064814805984497,0.7036458849906921,0.9990740418434143],[0.46250003576278687,0.3166666626930237,0.7052083760499954,0.9990740418434143],[0.4531250298023224,0.31296294927597046,0.6989583820104599,0.9990740418434143],[0.44322919845581055,0.31018519401550293,0.6911458820104599,0.9990740418434143],[0.4322916865348816,0.31388887763023376,0.6812500357627869,0.9990740716457367],[0.41979169845581055,0.31111109256744385,0.6703125536441803,0.9990740418434143],[0.3958333432674408,0.3157407343387604,0.6447916924953461,0.9990740716457367]]}]}],"rmot_expression_count":2,"rmot_negative_count":0,"rmot_motion_count":0}}}}},{"index":18,"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_011_scene_010.mp4","status":"complete","steps":{"decode":{"inputs":{"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_011_scene_010.mp4","sample_fps":4.0},"outputs":{"width":1920,"height":1080,"source_fps":50.0,"duration_seconds":16.5,"frames_kept":64,"blank_frames_dropped":0,"timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88,10.14,10.4,10.66,10.92,11.18,11.44,11.700000000000001,11.96,12.22,12.48,12.74,13.0,13.26,13.52,13.780000000000001,14.040000000000001,14.3,14.56,14.82,15.08,15.34,15.6,15.860000000000001,16.12,16.38],"error":""}},"propose":{"inputs":{"model":"gemini-3.5-flash-lite","temperature":0.4,"video_bytes":30727427,"system_prompt":"You are annotating video clips to build training data for referring multi-object tracking: given a video and a natural-language description, a model must output the boxes of every object matching that description, over time.\n\nYour job is to write the descriptions. You do NOT output coordinates, boxes, pixel positions, or any numbers about where things are. A separate detector finds the objects.\n\nFor every clip you produce a list of items. Each item has two texts that must NOT be the same:\n\nEXPRESSION — what a person would say to single out the object or objects. It may describe motion, temporal order, interaction with other objects, and spatial relations. It should read like something a person would actually say.\n  Good: \"the black car that's turning left\"\n  Good: \"the cyclist who overtakes the bus\"\n  Good: \"the two people who walk out of frame on the right\"\n\nNOUN_PHRASE — the same object as a plain, timeless visual category with its visible attributes, and NOTHING else. The detector is prompted with this text and looks at ONE frame at a time, so anything about time, order, or action is useless to it and makes it fail.\n  Allowed: category nouns, colours, materials, clothing, size, type\n  Forbidden: verbs and participles of action (\"turning\", \"walks\", \"is moving\"), temporal words (\"then\", \"starts\", \"after\", \"first\", \"last\"), relative clauses (\"that ...\", \"which ...\", \"who ...\"), ordinals and rank words (\"first\", \"leftmost\", \"nearest\"), and anything over six words.\n  \"the black car that's turning left\" -> \"black car\"\n  \"the cyclist who overtakes the bus\" -> \"cyclist\"\n  \"the two people who walk out of frame on the right\" -> \"person\"\n\nKIND — what makes the expression hard: \"motion\", \"appearance\", \"spatial\", or \"interaction\". Aim for about 35% \"motion\" items: motion is the part a single-frame model cannot already solve, so it is the most valuable, but a list of only motion items is not usable training data either.\n\nEXPECTED_COUNT — how many distinct objects in the clip match the EXPRESSION. Count carefully; this is checked against the detector's answer, and a wrong count discards the item.\n\nIS_NEGATIVE — at most ONE item may be a negative, and only when you also write at least one non-negative item. A negative is an expression that recombines attributes that ARE present in the clip into a whole that is NOT. If the only black car is moving, \"the parked black car\" is a good negative. Its NOUN_PHRASE must still be the category that IS present (\"black car\"), because that is what makes the absence checkable rather than merely unverified. A negative's EXPECTED_COUNT is 0.\n\nRules that discard an item, so follow them exactly:\n- Never write coordinates, box values, pixel positions, or percentages of the frame.\n- Never let a NOUN_PHRASE contain a verb of action, a temporal word, a relative clause, or an ordinal.\n- Never write an EXPRESSION whose object is not visible in the clip, unless it is marked IS_NEGATIVE.\n- Prefer objects that are clearly visible and identifiable over background clutter.\n- Do not describe the camera or the shot; describe objects in the scene.\n","user_prompt":"Watch this clip and write between 1 and 3 items following the rules. Cover different objects rather than restating one object several ways, and include any negative in the same list.\n"},"outputs":{"raw_response":"{\n  \"items\": [\n    {\n      \"expression\": \"the white pickup truck parked in the narrow alley\",\n      \"noun_phrase\": \"white pickup truck\",\n      \"kind\": \"spatial\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    }\n  ]\n}","raw_items":[{"expression":"the white pickup truck parked in the narrow alley","noun_phrase":"white pickup truck","kind":"spatial","expected_count":1,"is_negative":false}],"accepted":[{"expression":"the white pickup truck parked in the narrow alley","noun_phrase":"white pickup truck","kind":"spatial","expected_count":1,"is_negative":false}],"gated":[]}},"track":{"inputs":{"tracker":"Sam31MultiplexTracker","model_id":"/mnt/ceph3/ec/jacky/sam3.1","noun_phrases":["white pickup truck"],"expressions_in":1,"frames":64,"sample_fps":4.0,"frame_size":"1920x1080","timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88,10.14,10.4,10.66,10.92,11.18,11.44,11.700000000000001,11.96,12.22,12.48,12.74,13.0,13.26,13.52,13.780000000000001,14.040000000000001,14.3,14.56,14.82,15.08,15.34,15.6,15.860000000000001,16.12,16.38],"session":"one session over 64 JPEG frames; then per phrase: reset_session, add_prompt(text) on frame 0, propagate_in_video","passes":1},"outputs":{"raw_tracks":[{"noun_phrase":"white pickup truck","object_id":0,"observations":[{"timestamp":9.1,"box":[0.16302084922790527,0.44722220301628113,0.6968750357627869,0.7833333015441895],"occluded":false},{"timestamp":9.36,"box":[0.21302084624767303,0.46574074029922485,0.7713541835546494,0.8361110985279083],"occluded":false},{"timestamp":9.620000000000001,"box":[0.2880208492279053,0.44814813137054443,0.8776041865348816,0.8537036776542664],"occluded":false},{"timestamp":9.88,"box":[0.4333333671092987,0.5185185074806213,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":10.14,"box":[0.47083336114883423,0.5694444179534912,0.9994792342185974,0.9990740418434143],"occluded":false},{"timestamp":10.4,"box":[0.4776041805744171,0.5046296119689941,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":10.66,"box":[0.32239586114883423,0.5018518567085266,0.9994792342185974,0.970370352268219],"occluded":false},{"timestamp":10.92,"box":[0.05312500149011612,0.470370352268219,0.7010416761040688,0.9638888537883759],"occluded":false},{"timestamp":11.18,"box":[0.0,0.520370364189148,0.5286458730697632,0.9944444298744202],"occluded":false},{"timestamp":11.44,"box":[0.0,0.47870370745658875,0.4046875238418579,0.9740740656852722],"occluded":false},{"timestamp":11.700000000000001,"box":[0.0,0.34074074029922485,0.23906251788139343,0.9731481075286865],"occluded":false},{"timestamp":11.96,"box":[0.0,0.46481481194496155,0.1510416716337204,0.9759258925914764],"occluded":false},{"timestamp":12.22,"box":[0.0,0.5462962985038757,0.07968750596046448,0.9509259164333344],"occluded":false}],"scores":[0.7575758099555969,0.7575758099555969,0.7575758099555969,0.7575758099555969,0.7575758099555969,0.7575758099555969,0.7575758099555969,0.7575758099555969,0.7575758099555969,0.7575758099555969,0.7575758099555969,0.7575758099555969,0.7575758099555969]}],"tracklets":[{"noun_phrase":"white pickup truck","object_id":0,"presentation_id":1,"observations":[{"timestamp":9.1,"box":[0.16302084922790527,0.44722220301628113,0.6968750357627869,0.7833333015441895],"occluded":false},{"timestamp":9.36,"box":[0.21302084624767303,0.46574074029922485,0.7713541835546494,0.8361110985279083],"occluded":false},{"timestamp":9.620000000000001,"box":[0.2880208492279053,0.44814813137054443,0.8776041865348816,0.8537036776542664],"occluded":false},{"timestamp":9.88,"box":[0.4333333671092987,0.5185185074806213,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":10.14,"box":[0.47083336114883423,0.5694444179534912,0.9994792342185974,0.9990740418434143],"occluded":false},{"timestamp":10.4,"box":[0.4776041805744171,0.5046296119689941,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":10.66,"box":[0.32239586114883423,0.5018518567085266,0.9994792342185974,0.970370352268219],"occluded":false},{"timestamp":10.92,"box":[0.05312500149011612,0.470370352268219,0.7010416761040688,0.9638888537883759],"occluded":false},{"timestamp":11.18,"box":[0.0,0.520370364189148,0.5286458730697632,0.9944444298744202],"occluded":false},{"timestamp":11.44,"box":[0.0,0.47870370745658875,0.4046875238418579,0.9740740656852722],"occluded":false},{"timestamp":11.700000000000001,"box":[0.0,0.34074074029922485,0.23906251788139343,0.9731481075286865],"occluded":false},{"timestamp":11.96,"box":[0.0,0.46481481194496155,0.1510416716337204,0.9759258925914764],"occluded":false},{"timestamp":12.22,"box":[0.0,0.5462962985038757,0.07968750596046448,0.9509259164333344],"occluded":false}],"mean_score":0.7575758099555969}],"dropped":{},"fates":[{"noun_phrase":"white pickup truck","object_id":0,"frames":13,"mean_score":0.7575758099555969,"mean_area":0.20784945087048315,"outcome":"kept","presentation_id":1,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null}],"gates":{"min_track_frames":3,"min_mean_score":0.5,"merge_iou":0.8,"max_tracklets":12}}},"render":{"inputs":{"tracklets":1,"as_video":true,"fps":1.0,"max_frames":30},"outputs":{"media":"/home/zzk/data/video_bbox/vlm_gemini3.5_lite/renders/--cif7Nv6BI_chunk_011_scene_010_choices.mp4","media_bytes":5217427,"frames":[{"index":1,"timestamp":0.0,"file":"--cif7Nv6BI_chunk_011_scene_010_choices.mp4"},{"index":2,"timestamp":1.04,"file":"--cif7Nv6BI_chunk_011_scene_010_choices.mp4"},{"index":3,"timestamp":2.08,"file":"--cif7Nv6BI_chunk_011_scene_010_choices.mp4"},{"index":4,"timestamp":3.12,"file":"--cif7Nv6BI_chunk_011_scene_010_choices.mp4"},{"index":5,"timestamp":4.16,"file":"--cif7Nv6BI_chunk_011_scene_010_choices.mp4"},{"index":6,"timestamp":5.2,"file":"--cif7Nv6BI_chunk_011_scene_010_choices.mp4"},{"index":7,"timestamp":6.24,"file":"--cif7Nv6BI_chunk_011_scene_010_choices.mp4"},{"index":8,"timestamp":7.28,"file":"--cif7Nv6BI_chunk_011_scene_010_choices.mp4"},{"index":9,"timestamp":8.32,"file":"--cif7Nv6BI_chunk_011_scene_010_choices.mp4"},{"index":10,"timestamp":9.1,"file":"--cif7Nv6BI_chunk_011_scene_010_choices.mp4"},{"index":11,"timestamp":10.14,"file":"--cif7Nv6BI_chunk_011_scene_010_choices.mp4"},{"index":12,"timestamp":11.18,"file":"--cif7Nv6BI_chunk_011_scene_010_choices.mp4"},{"index":13,"timestamp":12.22,"file":"--cif7Nv6BI_chunk_011_scene_010_choices.mp4"},{"index":14,"timestamp":13.26,"file":"--cif7Nv6BI_chunk_011_scene_010_choices.mp4"},{"index":15,"timestamp":14.3,"file":"--cif7Nv6BI_chunk_011_scene_010_choices.mp4"},{"index":16,"timestamp":15.34,"file":"--cif7Nv6BI_chunk_011_scene_010_choices.mp4"},{"index":17,"timestamp":16.38,"file":"--cif7Nv6BI_chunk_011_scene_010_choices.mp4"}]}},"select":{"inputs":{"model":"gemini-3.5-flash-lite","temperature":0.0,"as_video":true,"media_bytes":5217427,"media":["--cif7Nv6BI_chunk_011_scene_010_choices.mp4"],"choice_frames":17,"system_prompt":"You are labelling video data for referring multi-object tracking.\n\nYou will see a numbered sequence of frames from one clip, in time order. Each frame has its own number and timestamp printed in the top-left corner. Candidate objects have been detected for you and drawn on the frames as coloured boxes, each labelled with a number. The same object keeps the same number and the same colour on every frame. Below each frame is a strip of enlarged crops of those same numbered boxes, so you can identify a small or distant object.\n\nFor each description you are given, answer two questions:\n\n1. MATCHING_IDS — which of the numbered objects the description refers to. List every number that matches, or an EMPTY LIST if none of them do. An empty list is a correct and expected answer: some descriptions are deliberately about something that is not in this clip. Do not pick a number just to have picked one, and do not pick a number for an object that merely resembles the description.\n\n2. START_FRAME and END_FRAME — the frame numbers between which the described state actually holds. If the description is \"the car that's turning left\", the interval is the frames in which that car is turning, NOT every frame in which the car is visible. Use the frame numbers printed on the frames. START_FRAME must not be greater than END_FRAME. When MATCHING_IDS is empty, give the full range (the first and last frame numbers); it is ignored.\n\n3. CONFIDENCE — how sure you are of the MATCHING_IDS answer, from 0.0 to 1.0. Be honest: a low number costs the example nothing, and an overconfident wrong answer becomes a wrong label.\n\nRules:\n- Only use numbers that are actually drawn on the frames.\n- Never output coordinates, box values, or pixel positions.\n- Judge motion by comparing the object's position across frames, not by guessing from one frame.\n","user_prompt":"The video has 17 frames, numbered 1 to 17. Each frame is one moment of the clip and has its number burned into the top left corner.\n\nAnswer for each of the following descriptions, in the order given:\n1. the white pickup truck parked in the narrow alley","questions":["the white pickup truck parked in the narrow alley"]},"outputs":{"raw_response":"{\n  \"answers\": [\n    {\n      \"description_index\": 1,\n      \"matching_ids\": [1],\n      \"start_frame\": 1,\n      \"end_frame\": 13,\n      \"confidence\": 1.0\n    }\n  ]\n}","raw_answers":[{"description_index":1,"matching_ids":[1],"start_frame":1,"end_frame":13,"confidence":1.0}],"answers":[{"expression":"the white pickup truck parked in the narrow alley","matching_presentation_ids":[1],"start_choice_frame":1,"end_choice_frame":13,"confidence":1.0,"self_reported_confidence":1.0}]}},"verify":{"inputs":{"expressions":1,"tracklets":1,"selections":1,"gates":"interval, count, unknown ids, confidence, then geometry (motion)"},"outputs":{"kept":[{"expression":"the white pickup truck parked in the narrow alley","kind":"spatial","is_negative":false,"tracks":[{"timestamps":[9.1,9.36,9.620000000000001,9.88,10.14,10.4,10.66,10.92,11.18,11.44,11.700000000000001,11.96,12.22],"boxes":[[0.16302084922790527,0.44722220301628113,0.6968750357627869,0.7833333015441895],[0.21302084624767303,0.46574074029922485,0.7713541835546494,0.8361110985279083],[0.2880208492279053,0.44814813137054443,0.8776041865348816,0.8537036776542664],[0.4333333671092987,0.5185185074806213,0.999479204416275,0.9722222089767456],[0.47083336114883423,0.5694444179534912,0.9994792342185974,0.9990740418434143],[0.4776041805744171,0.5046296119689941,0.999479204416275,0.9722222089767456],[0.32239586114883423,0.5018518567085266,0.9994792342185974,0.970370352268219],[0.05312500149011612,0.470370352268219,0.7010416761040688,0.9638888537883759],[0.0,0.520370364189148,0.5286458730697632,0.9944444298744202],[0.0,0.47870370745658875,0.4046875238418579,0.9740740656852722],[0.0,0.34074074029922485,0.23906251788139343,0.9731481075286865],[0.0,0.46481481194496155,0.1510416716337204,0.9759258925914764],[0.0,0.5462962985038757,0.07968750596046448,0.9509259164333344]]}]}],"rejected":{},"rejection_counts":{},"row":{"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_011_scene_010.mp4","scene_key":"--cif7Nv6BI_chunk_011_scene_010","category":"test_vehicles","height":1080,"width":1920,"fps":50.0,"num_frames":825,"duration":16.5,"rmot_expressions":[{"expression":"the white pickup truck parked in the narrow alley","kind":"spatial","is_negative":false,"tracks":[{"timestamps":[9.1,9.36,9.620000000000001,9.88,10.14,10.4,10.66,10.92,11.18,11.44,11.700000000000001,11.96,12.22],"boxes":[[0.16302084922790527,0.44722220301628113,0.6968750357627869,0.7833333015441895],[0.21302084624767303,0.46574074029922485,0.7713541835546494,0.8361110985279083],[0.2880208492279053,0.44814813137054443,0.8776041865348816,0.8537036776542664],[0.4333333671092987,0.5185185074806213,0.999479204416275,0.9722222089767456],[0.47083336114883423,0.5694444179534912,0.9994792342185974,0.9990740418434143],[0.4776041805744171,0.5046296119689941,0.999479204416275,0.9722222089767456],[0.32239586114883423,0.5018518567085266,0.9994792342185974,0.970370352268219],[0.05312500149011612,0.470370352268219,0.7010416761040688,0.9638888537883759],[0.0,0.520370364189148,0.5286458730697632,0.9944444298744202],[0.0,0.47870370745658875,0.4046875238418579,0.9740740656852722],[0.0,0.34074074029922485,0.23906251788139343,0.9731481075286865],[0.0,0.46481481194496155,0.1510416716337204,0.9759258925914764],[0.0,0.5462962985038757,0.07968750596046448,0.9509259164333344]]}]}],"rmot_expression_count":1,"rmot_negative_count":0,"rmot_motion_count":0}}}}},{"index":19,"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_012_scene_020.mp4","status":"complete","steps":{"decode":{"inputs":{"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_012_scene_020.mp4","sample_fps":4.0},"outputs":{"width":1920,"height":1080,"source_fps":50.0,"duration_seconds":14.56,"frames_kept":57,"blank_frames_dropped":0,"timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88,10.14,10.4,10.66,10.92,11.18,11.44,11.700000000000001,11.96,12.22,12.48,12.74,13.0,13.26,13.52,13.780000000000001,14.040000000000001,14.3,14.56],"error":""}},"propose":{"inputs":{"model":"gemini-3.5-flash-lite","temperature":0.4,"video_bytes":13891853,"system_prompt":"You are annotating video clips to build training data for referring multi-object tracking: given a video and a natural-language description, a model must output the boxes of every object matching that description, over time.\n\nYour job is to write the descriptions. You do NOT output coordinates, boxes, pixel positions, or any numbers about where things are. A separate detector finds the objects.\n\nFor every clip you produce a list of items. Each item has two texts that must NOT be the same:\n\nEXPRESSION — what a person would say to single out the object or objects. It may describe motion, temporal order, interaction with other objects, and spatial relations. It should read like something a person would actually say.\n  Good: \"the black car that's turning left\"\n  Good: \"the cyclist who overtakes the bus\"\n  Good: \"the two people who walk out of frame on the right\"\n\nNOUN_PHRASE — the same object as a plain, timeless visual category with its visible attributes, and NOTHING else. The detector is prompted with this text and looks at ONE frame at a time, so anything about time, order, or action is useless to it and makes it fail.\n  Allowed: category nouns, colours, materials, clothing, size, type\n  Forbidden: verbs and participles of action (\"turning\", \"walks\", \"is moving\"), temporal words (\"then\", \"starts\", \"after\", \"first\", \"last\"), relative clauses (\"that ...\", \"which ...\", \"who ...\"), ordinals and rank words (\"first\", \"leftmost\", \"nearest\"), and anything over six words.\n  \"the black car that's turning left\" -> \"black car\"\n  \"the cyclist who overtakes the bus\" -> \"cyclist\"\n  \"the two people who walk out of frame on the right\" -> \"person\"\n\nKIND — what makes the expression hard: \"motion\", \"appearance\", \"spatial\", or \"interaction\". Aim for about 35% \"motion\" items: motion is the part a single-frame model cannot already solve, so it is the most valuable, but a list of only motion items is not usable training data either.\n\nEXPECTED_COUNT — how many distinct objects in the clip match the EXPRESSION. Count carefully; this is checked against the detector's answer, and a wrong count discards the item.\n\nIS_NEGATIVE — at most ONE item may be a negative, and only when you also write at least one non-negative item. A negative is an expression that recombines attributes that ARE present in the clip into a whole that is NOT. If the only black car is moving, \"the parked black car\" is a good negative. Its NOUN_PHRASE must still be the category that IS present (\"black car\"), because that is what makes the absence checkable rather than merely unverified. A negative's EXPECTED_COUNT is 0.\n\nRules that discard an item, so follow them exactly:\n- Never write coordinates, box values, pixel positions, or percentages of the frame.\n- Never let a NOUN_PHRASE contain a verb of action, a temporal word, a relative clause, or an ordinal.\n- Never write an EXPRESSION whose object is not visible in the clip, unless it is marked IS_NEGATIVE.\n- Prefer objects that are clearly visible and identifiable over background clutter.\n- Do not describe the camera or the shot; describe objects in the scene.\n","user_prompt":"Watch this clip and write between 1 and 3 items following the rules. Cover different objects rather than restating one object several ways, and include any negative in the same list.\n"},"outputs":{"raw_response":"{\n  \"items\": [\n    {\n      \"expression\": \"the glowing street lamp illuminating the narrow alleyway\",\n      \"noun_phrase\": \"street lamp\",\n      \"kind\": \"appearance\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the dark wooden door set into the cream-colored building\",\n      \"noun_phrase\": \"wooden door\",\n      \"kind\": \"spatial\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the modern red car parked along the stone alley\",\n      \"noun_phrase\": \"red car\",\n      \"kind\": \"appearance\",\n      \"expected_count\": 0,\n      \"is_negative\": true\n    }\n  ]\n}","raw_items":[{"expression":"the glowing street lamp illuminating the narrow alleyway","noun_phrase":"street lamp","kind":"appearance","expected_count":1,"is_negative":false},{"expression":"the dark wooden door set into the cream-colored building","noun_phrase":"wooden door","kind":"spatial","expected_count":1,"is_negative":false},{"expression":"the modern red car parked along the stone alley","noun_phrase":"red car","kind":"appearance","expected_count":0,"is_negative":true}],"accepted":[{"expression":"the glowing street lamp illuminating the narrow alleyway","noun_phrase":"street lamp","kind":"appearance","expected_count":1,"is_negative":false},{"expression":"the dark wooden door set into the cream-colored building","noun_phrase":"wooden door","kind":"spatial","expected_count":1,"is_negative":false},{"expression":"the modern red car parked along the stone alley","noun_phrase":"red car","kind":"appearance","expected_count":0,"is_negative":true}],"gated":[]}},"track":{"inputs":{"tracker":"Sam31MultiplexTracker","model_id":"/mnt/ceph3/ec/jacky/sam3.1","noun_phrases":["red car","street lamp","wooden door"],"expressions_in":3,"frames":57,"sample_fps":4.0,"frame_size":"1920x1080","timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88,10.14,10.4,10.66,10.92,11.18,11.44,11.700000000000001,11.96,12.22,12.48,12.74,13.0,13.26,13.52,13.780000000000001,14.040000000000001,14.3,14.56],"session":"one session over 57 JPEG frames; then per phrase: reset_session, add_prompt(text) on frame 0, propagate_in_video","passes":3},"outputs":{"raw_tracks":[{"noun_phrase":"street lamp","object_id":0,"observations":[{"timestamp":0.0,"box":[0.5203125476837158,0.49537035822868347,0.5307292146608233,0.5185185056179762],"occluded":false},{"timestamp":0.26,"box":[0.5401042103767395,0.5481481552124023,0.5515625439584255,0.5712963026016951],"occluded":false},{"timestamp":0.52,"box":[0.6041666865348816,0.6120370030403137,0.6213541869074106,0.6361110769212246],"occluded":false},{"timestamp":0.78,"box":[0.7791666984558105,0.7379629611968994,0.8072916995733976,0.7685185167938471],"occluded":false},{"timestamp":4.16,"box":[0.8276042342185974,0.6296296119689941,0.9901042431592941,0.8148147910833359],"occluded":false},{"timestamp":4.42,"box":[0.8265625238418579,0.6620370149612427,0.848958358168602,0.690740717574954],"occluded":false},{"timestamp":4.68,"box":[0.7401041984558105,0.5777777433395386,0.7687500324100256,0.6370370015501976],"occluded":false},{"timestamp":4.94,"box":[0.6692708730697632,0.49444442987442017,0.6828125407919288,0.5231481324881315],"occluded":false},{"timestamp":5.2,"box":[0.6130208373069763,0.4342592656612396,0.6244791708886623,0.45925926603376865],"occluded":false},{"timestamp":5.46,"box":[0.5838541984558105,0.4138888716697693,0.5953125320374966,0.43796294555068016],"occluded":false},{"timestamp":5.72,"box":[0.5750000476837158,0.4000000059604645,0.5869792150333524,0.4250000063329935],"occluded":false},{"timestamp":5.98,"box":[0.5536458492279053,0.4175925850868225,0.5651041828095913,0.4407407324761152],"occluded":false},{"timestamp":6.24,"box":[0.5442708730697632,0.4444444477558136,0.5546875400468707,0.46666666865348816],"occluded":false},{"timestamp":6.5,"box":[0.5427083373069763,0.43518516421318054,0.5531250042840838,0.4592592380940914],"occluded":false},{"timestamp":6.76,"box":[0.5354167222976685,0.48055553436279297,0.5468750558793545,0.505555534735322],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.534375011920929,0.4814814627170563,0.545833345502615,0.5064814630895853],"occluded":false},{"timestamp":7.28,"box":[0.5338541865348816,0.44814813137054443,0.5447916872799397,0.470370352268219],"occluded":false},{"timestamp":7.54,"box":[0.5223958492279053,0.4611110985279083,0.5380208492279053,0.7962962687015533],"occluded":false},{"timestamp":7.8,"box":[0.5197917222976685,0.45370370149612427,0.5307292230427265,0.4787037018686533],"occluded":false},{"timestamp":8.06,"box":[0.5458333492279053,0.45277777314186096,0.5572916828095913,0.4824074022471905],"occluded":false},{"timestamp":8.32,"box":[0.5359375476837158,0.489814817905426,0.5531250480562449,0.8518518507480621],"occluded":false},{"timestamp":8.58,"box":[0.5359375476837158,0.46574074029922485,0.5520833823829889,0.8379629552364349],"occluded":false},{"timestamp":8.84,"box":[0.5338541865348816,0.43981480598449707,0.5458333538845181,0.46666665747761726],"occluded":false},{"timestamp":9.1,"box":[0.5364583730697632,0.42129629850387573,0.5536458734422922,0.8101851940155029],"occluded":false},{"timestamp":9.36,"box":[0.5479167103767395,0.4000000059604645,0.5604167114943266,0.7962962985038757],"occluded":false},{"timestamp":9.620000000000001,"box":[0.5473958849906921,0.41111111640930176,0.5604167189449072,0.8240740597248077],"occluded":false},{"timestamp":9.88,"box":[0.5453125238418579,0.4055555462837219,0.558333357796073,0.8379629552364349],"occluded":false},{"timestamp":10.14,"box":[0.5390625,0.40925925970077515,0.5604166686534882,0.8592592477798462],"occluded":false},{"timestamp":10.4,"box":[0.5255208611488342,0.42314815521240234,0.5468750298023224,0.8870370388031006],"occluded":false},{"timestamp":10.66,"box":[0.5208333730697632,0.4157407283782959,0.5604167096316814,0.9009259045124054],"occluded":false},{"timestamp":10.92,"box":[0.5255208611488342,0.42129629850387573,0.5609375312924385,0.9425925612449646],"occluded":false},{"timestamp":11.18,"box":[0.5302083492279053,0.4259259104728699,0.5744791850447655,0.970370352268219],"occluded":false},{"timestamp":11.44,"box":[0.5395833849906921,0.4194444417953491,0.5536458855494857,0.4481481444090605],"occluded":false},{"timestamp":11.700000000000001,"box":[0.5416666865348816,0.43518516421318054,0.5557291870936751,0.46481479331851006],"occluded":false},{"timestamp":11.96,"box":[0.5406250357627869,0.4333333373069763,0.5546875363215804,0.46296296641230583],"occluded":false},{"timestamp":12.22,"box":[0.5447916984558105,0.4185185134410858,0.5598958656191826,0.44999999552965164],"occluded":false},{"timestamp":12.48,"box":[0.5484375357627869,0.4287036955356598,0.5625000363215804,0.4601851776242256],"occluded":false},{"timestamp":12.74,"box":[0.5526041984558105,0.42499998211860657,0.5666666990146041,0.4564814642071724],"occluded":false},{"timestamp":13.0,"box":[0.5520833730697632,0.41111111640930176,0.5661458736285567,0.4425925984978676],"occluded":false},{"timestamp":13.26,"box":[0.5541666746139526,0.41018518805503845,0.5692708417773247,0.4435185194015503],"occluded":false},{"timestamp":13.52,"box":[0.5541666746139526,0.3898147940635681,0.5708333421498537,0.42407405376434326],"occluded":false},{"timestamp":13.780000000000001,"box":[0.5640625357627869,0.4185185134410858,0.5796875357627869,0.45277777314186096],"occluded":false},{"timestamp":14.040000000000001,"box":[0.5781250596046448,0.4009259045124054,0.5942708943039179,0.43611108884215355],"occluded":false},{"timestamp":14.3,"box":[0.5765625238418579,0.3907407224178314,0.592708358541131,0.4259259067475796],"occluded":false},{"timestamp":14.56,"box":[0.5640625357627869,0.402777761220932,0.5807292032986879,0.43888887017965317],"occluded":false}],"scores":[0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295,0.6908212304115295]},{"noun_phrase":"street lamp","object_id":1,"observations":[{"timestamp":0.0,"box":[0.6432291865348816,0.031481482088565826,0.7651041895151138,0.17314814776182175],"occluded":false},{"timestamp":0.26,"box":[0.6619791984558105,0.043518517166376114,0.7927083671092987,0.21574072912335396],"occluded":false},{"timestamp":0.52,"box":[0.7119792103767395,0.069444440305233,0.8651042133569717,0.2212962880730629],"occluded":false},{"timestamp":0.78,"box":[0.7828125357627869,0.08703703433275223,0.9994792193174362,0.7407407239079475],"occluded":false},{"timestamp":1.04,"box":[0.9765625596046448,0.09444444626569748,0.9994792267680168,0.18703703582286835],"occluded":false}],"scores":[0.9064327478408813,0.9064327478408813,0.9064327478408813,0.9064327478408813,0.9064327478408813]},{"noun_phrase":"wooden door","object_id":0,"observations":[{"timestamp":4.42,"box":[0.0,0.25740739703178406,0.29114586114883423,0.9703703224658966],"occluded":false},{"timestamp":4.68,"box":[0.0,0.1944444477558136,0.18645834922790527,0.9694444239139557],"occluded":false},{"timestamp":4.94,"box":[0.0,0.2666666507720947,0.02291666716337204,0.416666641831398],"occluded":false}],"scores":[0.6800000071525574,0.6800000071525574,0.6800000071525574]},{"noun_phrase":"wooden door","object_id":1,"observations":[{"timestamp":4.42,"box":[0.5958333611488342,0.7712962627410889,0.6265625301748514,0.9703703373670578],"occluded":false},{"timestamp":4.68,"box":[0.5145833492279053,0.67037034034729,0.5552083514630795,0.9398147761821747],"occluded":false},{"timestamp":4.94,"box":[0.4541666805744171,0.5972222089767456,0.48906251415610313,0.8287036865949631],"occluded":false},{"timestamp":5.2,"box":[0.39010417461395264,0.5351851582527161,0.4296875111758709,0.7703703343868256],"occluded":false},{"timestamp":5.46,"box":[0.34843751788139343,0.5037037134170532,0.3942708522081375,0.7546296417713165],"occluded":false},{"timestamp":5.72,"box":[0.32864585518836975,0.4833333194255829,0.3776041902601719,0.7416666448116302],"occluded":false},{"timestamp":5.98,"box":[0.2979166805744171,0.4972222149372101,0.34947918355464935,0.7740740478038788],"occluded":false},{"timestamp":6.24,"box":[0.27031251788139343,0.5129629373550415,0.3307291865348816,0.8074073791503906],"occluded":false},{"timestamp":6.5,"box":[0.25416669249534607,0.49537035822868347,0.3213541954755783,0.8055555522441864],"occluded":false},{"timestamp":6.76,"box":[0.22447918355464935,0.5527777671813965,0.3005208522081375,0.895370364189148],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.20468750596046448,0.5564814805984497,0.28958334028720856,0.9157407283782959],"occluded":false},{"timestamp":7.28,"box":[0.19166667759418488,0.5157407522201538,0.2802083492279053,0.9009259343147278],"occluded":false},{"timestamp":7.54,"box":[0.15312500298023224,0.5185185074806213,0.2593750059604645,0.9564814567565918],"occluded":false},{"timestamp":7.8,"box":[0.11510417610406876,0.5037037134170532,0.2395833507180214,0.9620370268821716],"occluded":false},{"timestamp":8.06,"box":[0.11250000447034836,0.4824073910713196,0.25468751043081284,0.970370352268219],"occluded":false}],"scores":[0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563]}],"tracklets":[{"noun_phrase":"street lamp","object_id":0,"presentation_id":1,"observations":[{"timestamp":0.0,"box":[0.5203125476837158,0.49537035822868347,0.5307292146608233,0.5185185056179762],"occluded":false},{"timestamp":0.26,"box":[0.5401042103767395,0.5481481552124023,0.5515625439584255,0.5712963026016951],"occluded":false},{"timestamp":0.52,"box":[0.6041666865348816,0.6120370030403137,0.6213541869074106,0.6361110769212246],"occluded":false},{"timestamp":0.78,"box":[0.7791666984558105,0.7379629611968994,0.8072916995733976,0.7685185167938471],"occluded":false},{"timestamp":4.16,"box":[0.8276042342185974,0.6296296119689941,0.9901042431592941,0.8148147910833359],"occluded":false},{"timestamp":4.42,"box":[0.8265625238418579,0.6620370149612427,0.848958358168602,0.690740717574954],"occluded":false},{"timestamp":4.68,"box":[0.7401041984558105,0.5777777433395386,0.7687500324100256,0.6370370015501976],"occluded":false},{"timestamp":4.94,"box":[0.6692708730697632,0.49444442987442017,0.6828125407919288,0.5231481324881315],"occluded":false},{"timestamp":5.2,"box":[0.6130208373069763,0.4342592656612396,0.6244791708886623,0.45925926603376865],"occluded":false},{"timestamp":5.46,"box":[0.5838541984558105,0.4138888716697693,0.5953125320374966,0.43796294555068016],"occluded":false},{"timestamp":5.72,"box":[0.5750000476837158,0.4000000059604645,0.5869792150333524,0.4250000063329935],"occluded":false},{"timestamp":5.98,"box":[0.5536458492279053,0.4175925850868225,0.5651041828095913,0.4407407324761152],"occluded":false},{"timestamp":6.24,"box":[0.5442708730697632,0.4444444477558136,0.5546875400468707,0.46666666865348816],"occluded":false},{"timestamp":6.5,"box":[0.5427083373069763,0.43518516421318054,0.5531250042840838,0.4592592380940914],"occluded":false},{"timestamp":6.76,"box":[0.5354167222976685,0.48055553436279297,0.5468750558793545,0.505555534735322],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.534375011920929,0.4814814627170563,0.545833345502615,0.5064814630895853],"occluded":false},{"timestamp":7.28,"box":[0.5338541865348816,0.44814813137054443,0.5447916872799397,0.470370352268219],"occluded":false},{"timestamp":7.54,"box":[0.5223958492279053,0.4611110985279083,0.5380208492279053,0.7962962687015533],"occluded":false},{"timestamp":7.8,"box":[0.5197917222976685,0.45370370149612427,0.5307292230427265,0.4787037018686533],"occluded":false},{"timestamp":8.06,"box":[0.5458333492279053,0.45277777314186096,0.5572916828095913,0.4824074022471905],"occluded":false},{"timestamp":8.32,"box":[0.5359375476837158,0.489814817905426,0.5531250480562449,0.8518518507480621],"occluded":false},{"timestamp":8.58,"box":[0.5359375476837158,0.46574074029922485,0.5520833823829889,0.8379629552364349],"occluded":false},{"timestamp":8.84,"box":[0.5338541865348816,0.43981480598449707,0.5458333538845181,0.46666665747761726],"occluded":false},{"timestamp":9.1,"box":[0.5364583730697632,0.42129629850387573,0.5536458734422922,0.8101851940155029],"occluded":false},{"timestamp":9.36,"box":[0.5479167103767395,0.4000000059604645,0.5604167114943266,0.7962962985038757],"occluded":false},{"timestamp":9.620000000000001,"box":[0.5473958849906921,0.41111111640930176,0.5604167189449072,0.8240740597248077],"occluded":false},{"timestamp":9.88,"box":[0.5453125238418579,0.4055555462837219,0.558333357796073,0.8379629552364349],"occluded":false},{"timestamp":10.14,"box":[0.5390625,0.40925925970077515,0.5604166686534882,0.8592592477798462],"occluded":false},{"timestamp":10.4,"box":[0.5255208611488342,0.42314815521240234,0.5468750298023224,0.8870370388031006],"occluded":false},{"timestamp":10.66,"box":[0.5208333730697632,0.4157407283782959,0.5604167096316814,0.9009259045124054],"occluded":false},{"timestamp":10.92,"box":[0.5255208611488342,0.42129629850387573,0.5609375312924385,0.9425925612449646],"occluded":false},{"timestamp":11.18,"box":[0.5302083492279053,0.4259259104728699,0.5744791850447655,0.970370352268219],"occluded":false},{"timestamp":11.44,"box":[0.5395833849906921,0.4194444417953491,0.5536458855494857,0.4481481444090605],"occluded":false},{"timestamp":11.700000000000001,"box":[0.5416666865348816,0.43518516421318054,0.5557291870936751,0.46481479331851006],"occluded":false},{"timestamp":11.96,"box":[0.5406250357627869,0.4333333373069763,0.5546875363215804,0.46296296641230583],"occluded":false},{"timestamp":12.22,"box":[0.5447916984558105,0.4185185134410858,0.5598958656191826,0.44999999552965164],"occluded":false},{"timestamp":12.48,"box":[0.5484375357627869,0.4287036955356598,0.5625000363215804,0.4601851776242256],"occluded":false},{"timestamp":12.74,"box":[0.5526041984558105,0.42499998211860657,0.5666666990146041,0.4564814642071724],"occluded":false},{"timestamp":13.0,"box":[0.5520833730697632,0.41111111640930176,0.5661458736285567,0.4425925984978676],"occluded":false},{"timestamp":13.26,"box":[0.5541666746139526,0.41018518805503845,0.5692708417773247,0.4435185194015503],"occluded":false},{"timestamp":13.52,"box":[0.5541666746139526,0.3898147940635681,0.5708333421498537,0.42407405376434326],"occluded":false},{"timestamp":13.780000000000001,"box":[0.5640625357627869,0.4185185134410858,0.5796875357627869,0.45277777314186096],"occluded":false},{"timestamp":14.040000000000001,"box":[0.5781250596046448,0.4009259045124054,0.5942708943039179,0.43611108884215355],"occluded":false},{"timestamp":14.3,"box":[0.5765625238418579,0.3907407224178314,0.592708358541131,0.4259259067475796],"occluded":false},{"timestamp":14.56,"box":[0.5640625357627869,0.402777761220932,0.5807292032986879,0.43888887017965317],"occluded":false}],"mean_score":0.6908212304115295},{"noun_phrase":"wooden door","object_id":0,"presentation_id":2,"observations":[{"timestamp":4.42,"box":[0.0,0.25740739703178406,0.29114586114883423,0.9703703224658966],"occluded":false},{"timestamp":4.68,"box":[0.0,0.1944444477558136,0.18645834922790527,0.9694444239139557],"occluded":false},{"timestamp":4.94,"box":[0.0,0.2666666507720947,0.02291666716337204,0.416666641831398],"occluded":false}],"mean_score":0.6800000071525574},{"noun_phrase":"street lamp","object_id":1,"presentation_id":3,"observations":[{"timestamp":0.0,"box":[0.6432291865348816,0.031481482088565826,0.7651041895151138,0.17314814776182175],"occluded":false},{"timestamp":0.26,"box":[0.6619791984558105,0.043518517166376114,0.7927083671092987,0.21574072912335396],"occluded":false},{"timestamp":0.52,"box":[0.7119792103767395,0.069444440305233,0.8651042133569717,0.2212962880730629],"occluded":false},{"timestamp":0.78,"box":[0.7828125357627869,0.08703703433275223,0.9994792193174362,0.7407407239079475],"occluded":false},{"timestamp":1.04,"box":[0.9765625596046448,0.09444444626569748,0.9994792267680168,0.18703703582286835],"occluded":false}],"mean_score":0.9064327478408813},{"noun_phrase":"wooden door","object_id":1,"presentation_id":4,"observations":[{"timestamp":4.42,"box":[0.5958333611488342,0.7712962627410889,0.6265625301748514,0.9703703373670578],"occluded":false},{"timestamp":4.68,"box":[0.5145833492279053,0.67037034034729,0.5552083514630795,0.9398147761821747],"occluded":false},{"timestamp":4.94,"box":[0.4541666805744171,0.5972222089767456,0.48906251415610313,0.8287036865949631],"occluded":false},{"timestamp":5.2,"box":[0.39010417461395264,0.5351851582527161,0.4296875111758709,0.7703703343868256],"occluded":false},{"timestamp":5.46,"box":[0.34843751788139343,0.5037037134170532,0.3942708522081375,0.7546296417713165],"occluded":false},{"timestamp":5.72,"box":[0.32864585518836975,0.4833333194255829,0.3776041902601719,0.7416666448116302],"occluded":false},{"timestamp":5.98,"box":[0.2979166805744171,0.4972222149372101,0.34947918355464935,0.7740740478038788],"occluded":false},{"timestamp":6.24,"box":[0.27031251788139343,0.5129629373550415,0.3307291865348816,0.8074073791503906],"occluded":false},{"timestamp":6.5,"box":[0.25416669249534607,0.49537035822868347,0.3213541954755783,0.8055555522441864],"occluded":false},{"timestamp":6.76,"box":[0.22447918355464935,0.5527777671813965,0.3005208522081375,0.895370364189148],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.20468750596046448,0.5564814805984497,0.28958334028720856,0.9157407283782959],"occluded":false},{"timestamp":7.28,"box":[0.19166667759418488,0.5157407522201538,0.2802083492279053,0.9009259343147278],"occluded":false},{"timestamp":7.54,"box":[0.15312500298023224,0.5185185074806213,0.2593750059604645,0.9564814567565918],"occluded":false},{"timestamp":7.8,"box":[0.11510417610406876,0.5037037134170532,0.2395833507180214,0.9620370268821716],"occluded":false},{"timestamp":8.06,"box":[0.11250000447034836,0.4824073910713196,0.25468751043081284,0.970370352268219],"occluded":false}],"mean_score":0.6559139490127563}],"dropped":{},"fates":[{"noun_phrase":"street lamp","object_id":0,"frames":45,"mean_score":0.6908212304115295,"mean_area":0.0036833420560621073,"outcome":"kept","presentation_id":1,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"street lamp","object_id":1,"frames":5,"mean_score":0.9064327478408813,"mean_area":0.041358026108060574,"outcome":"kept","presentation_id":3,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"wooden door","object_id":0,"frames":3,"mean_score":0.6800000071525574,"mean_area":0.11850630698947824,"outcome":"kept","presentation_id":2,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"wooden door","object_id":1,"frames":15,"mean_score":0.6559139490127563,"mean_area":0.02500858470415809,"outcome":"kept","presentation_id":4,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null}],"gates":{"min_track_frames":3,"min_mean_score":0.5,"merge_iou":0.8,"max_tracklets":12}}},"render":{"inputs":{"tracklets":4,"as_video":true,"fps":1.0,"max_frames":30},"outputs":{"media":"/home/zzk/data/video_bbox/vlm_gemini3.5_lite/renders/--cif7Nv6BI_chunk_012_scene_020_choices.mp4","media_bytes":3505288,"frames":[{"index":1,"timestamp":0.0,"file":"--cif7Nv6BI_chunk_012_scene_020_choices.mp4"},{"index":2,"timestamp":1.04,"file":"--cif7Nv6BI_chunk_012_scene_020_choices.mp4"},{"index":3,"timestamp":1.82,"file":"--cif7Nv6BI_chunk_012_scene_020_choices.mp4"},{"index":4,"timestamp":2.86,"file":"--cif7Nv6BI_chunk_012_scene_020_choices.mp4"},{"index":5,"timestamp":3.9,"file":"--cif7Nv6BI_chunk_012_scene_020_choices.mp4"},{"index":6,"timestamp":4.94,"file":"--cif7Nv6BI_chunk_012_scene_020_choices.mp4"},{"index":7,"timestamp":5.72,"file":"--cif7Nv6BI_chunk_012_scene_020_choices.mp4"},{"index":8,"timestamp":6.76,"file":"--cif7Nv6BI_chunk_012_scene_020_choices.mp4"},{"index":9,"timestamp":7.8,"file":"--cif7Nv6BI_chunk_012_scene_020_choices.mp4"},{"index":10,"timestamp":8.84,"file":"--cif7Nv6BI_chunk_012_scene_020_choices.mp4"},{"index":11,"timestamp":9.620000000000001,"file":"--cif7Nv6BI_chunk_012_scene_020_choices.mp4"},{"index":12,"timestamp":10.66,"file":"--cif7Nv6BI_chunk_012_scene_020_choices.mp4"},{"index":13,"timestamp":11.700000000000001,"file":"--cif7Nv6BI_chunk_012_scene_020_choices.mp4"},{"index":14,"timestamp":12.74,"file":"--cif7Nv6BI_chunk_012_scene_020_choices.mp4"},{"index":15,"timestamp":13.52,"file":"--cif7Nv6BI_chunk_012_scene_020_choices.mp4"},{"index":16,"timestamp":14.56,"file":"--cif7Nv6BI_chunk_012_scene_020_choices.mp4"}]}},"select":{"inputs":{"model":"gemini-3.5-flash-lite","temperature":0.0,"as_video":true,"media_bytes":3505288,"media":["--cif7Nv6BI_chunk_012_scene_020_choices.mp4"],"choice_frames":16,"system_prompt":"You are labelling video data for referring multi-object tracking.\n\nYou will see a numbered sequence of frames from one clip, in time order. Each frame has its own number and timestamp printed in the top-left corner. Candidate objects have been detected for you and drawn on the frames as coloured boxes, each labelled with a number. The same object keeps the same number and the same colour on every frame. Below each frame is a strip of enlarged crops of those same numbered boxes, so you can identify a small or distant object.\n\nFor each description you are given, answer two questions:\n\n1. MATCHING_IDS — which of the numbered objects the description refers to. List every number that matches, or an EMPTY LIST if none of them do. An empty list is a correct and expected answer: some descriptions are deliberately about something that is not in this clip. Do not pick a number just to have picked one, and do not pick a number for an object that merely resembles the description.\n\n2. START_FRAME and END_FRAME — the frame numbers between which the described state actually holds. If the description is \"the car that's turning left\", the interval is the frames in which that car is turning, NOT every frame in which the car is visible. Use the frame numbers printed on the frames. START_FRAME must not be greater than END_FRAME. When MATCHING_IDS is empty, give the full range (the first and last frame numbers); it is ignored.\n\n3. CONFIDENCE — how sure you are of the MATCHING_IDS answer, from 0.0 to 1.0. Be honest: a low number costs the example nothing, and an overconfident wrong answer becomes a wrong label.\n\nRules:\n- Only use numbers that are actually drawn on the frames.\n- Never output coordinates, box values, or pixel positions.\n- Judge motion by comparing the object's position across frames, not by guessing from one frame.\n","user_prompt":"The video has 16 frames, numbered 1 to 16. Each frame is one moment of the clip and has its number burned into the top left corner.\n\nAnswer for each of the following descriptions, in the order given:\n1. the glowing street lamp illuminating the narrow alleyway\n2. the dark wooden door set into the cream-colored building\n3. the modern red car parked along the stone alley","questions":["the glowing street lamp illuminating the narrow alleyway","the dark wooden door set into the cream-colored building","the modern red car parked along the stone alley"]},"outputs":{"raw_response":"{\n  \"answers\": [\n    {\n      \"description_index\": 1,\n      \"matching_ids\": [\n        1\n      ],\n      \"start_frame\": 1,\n      \"end_frame\": 16,\n      \"confidence\": 1.0\n    },\n    {\n      \"description_index\": 2,\n      \"matching_ids\": [\n        4\n      ],\n      \"start_frame\": 6,\n      \"end_frame\": 9,\n      \"confidence\": 1.0\n    },\n    {\n      \"description_index\": 3,\n      \"matching_ids\": [],\n      \"start_frame\": 1,\n      \"end_frame\": 16,\n      \"confidence\": 1.0\n    }\n  ]\n}","raw_answers":[{"description_index":1,"matching_ids":[1],"start_frame":1,"end_frame":16,"confidence":1.0},{"description_index":2,"matching_ids":[4],"start_frame":6,"end_frame":9,"confidence":1.0},{"description_index":3,"matching_ids":[],"start_frame":1,"end_frame":16,"confidence":1.0}],"answers":[{"expression":"the glowing street lamp illuminating the narrow alleyway","matching_presentation_ids":[1],"start_choice_frame":1,"end_choice_frame":16,"confidence":1.0,"self_reported_confidence":1.0},{"expression":"the dark wooden door set into the cream-colored building","matching_presentation_ids":[4],"start_choice_frame":6,"end_choice_frame":9,"confidence":1.0,"self_reported_confidence":1.0},{"expression":"the modern red car parked along the stone alley","matching_presentation_ids":[],"start_choice_frame":1,"end_choice_frame":16,"confidence":1.0,"self_reported_confidence":1.0}]}},"verify":{"inputs":{"expressions":3,"tracklets":4,"selections":3,"gates":"interval, count, unknown ids, confidence, then geometry (motion)"},"outputs":{"kept":[{"expression":"the dark wooden door set into the cream-colored building","kind":"spatial","is_negative":false,"tracks":[{"timestamps":[4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8],"boxes":[[0.4541666805744171,0.5972222089767456,0.48906251415610313,0.8287036865949631],[0.39010417461395264,0.5351851582527161,0.4296875111758709,0.7703703343868256],[0.34843751788139343,0.5037037134170532,0.3942708522081375,0.7546296417713165],[0.32864585518836975,0.4833333194255829,0.3776041902601719,0.7416666448116302],[0.2979166805744171,0.4972222149372101,0.34947918355464935,0.7740740478038788],[0.27031251788139343,0.5129629373550415,0.3307291865348816,0.8074073791503906],[0.25416669249534607,0.49537035822868347,0.3213541954755783,0.8055555522441864],[0.22447918355464935,0.5527777671813965,0.3005208522081375,0.895370364189148],[0.20468750596046448,0.5564814805984497,0.28958334028720856,0.9157407283782959],[0.19166667759418488,0.5157407522201538,0.2802083492279053,0.9009259343147278],[0.15312500298023224,0.5185185074806213,0.2593750059604645,0.9564814567565918],[0.11510417610406876,0.5037037134170532,0.2395833507180214,0.9620370268821716]]}]}],"rejected":{"the modern red car parked along the stone alley":"no_candidates_retrieved","the glowing street lamp illuminating the narrow alleyway":"implausible_box"},"rejection_counts":{"no_candidates_retrieved":1,"implausible_box":1},"row":{"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_012_scene_020.mp4","scene_key":"--cif7Nv6BI_chunk_012_scene_020","category":"test_vehicles","height":1080,"width":1920,"fps":50.0,"num_frames":728,"duration":14.56,"rmot_expressions":[{"expression":"the dark wooden door set into the cream-colored building","kind":"spatial","is_negative":false,"tracks":[{"timestamps":[4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8],"boxes":[[0.4541666805744171,0.5972222089767456,0.48906251415610313,0.8287036865949631],[0.39010417461395264,0.5351851582527161,0.4296875111758709,0.7703703343868256],[0.34843751788139343,0.5037037134170532,0.3942708522081375,0.7546296417713165],[0.32864585518836975,0.4833333194255829,0.3776041902601719,0.7416666448116302],[0.2979166805744171,0.4972222149372101,0.34947918355464935,0.7740740478038788],[0.27031251788139343,0.5129629373550415,0.3307291865348816,0.8074073791503906],[0.25416669249534607,0.49537035822868347,0.3213541954755783,0.8055555522441864],[0.22447918355464935,0.5527777671813965,0.3005208522081375,0.895370364189148],[0.20468750596046448,0.5564814805984497,0.28958334028720856,0.9157407283782959],[0.19166667759418488,0.5157407522201538,0.2802083492279053,0.9009259343147278],[0.15312500298023224,0.5185185074806213,0.2593750059604645,0.9564814567565918],[0.11510417610406876,0.5037037134170532,0.2395833507180214,0.9620370268821716]]}]}],"rmot_expression_count":1,"rmot_negative_count":0,"rmot_motion_count":0}}}}},{"index":20,"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_014_scene_018.mp4","status":"complete","steps":{"decode":{"inputs":{"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_014_scene_018.mp4","sample_fps":4.0},"outputs":{"width":1920,"height":1080,"source_fps":50.0,"duration_seconds":12.56,"frames_kept":49,"blank_frames_dropped":0,"timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88,10.14,10.4,10.66,10.92,11.18,11.44,11.700000000000001,11.96,12.22,12.48],"error":""}},"propose":{"inputs":{"model":"gemini-3.5-flash-lite","temperature":0.4,"video_bytes":14915301,"system_prompt":"You are annotating video clips to build training data for referring multi-object tracking: given a video and a natural-language description, a model must output the boxes of every object matching that description, over time.\n\nYour job is to write the descriptions. You do NOT output coordinates, boxes, pixel positions, or any numbers about where things are. A separate detector finds the objects.\n\nFor every clip you produce a list of items. Each item has two texts that must NOT be the same:\n\nEXPRESSION — what a person would say to single out the object or objects. It may describe motion, temporal order, interaction with other objects, and spatial relations. It should read like something a person would actually say.\n  Good: \"the black car that's turning left\"\n  Good: \"the cyclist who overtakes the bus\"\n  Good: \"the two people who walk out of frame on the right\"\n\nNOUN_PHRASE — the same object as a plain, timeless visual category with its visible attributes, and NOTHING else. The detector is prompted with this text and looks at ONE frame at a time, so anything about time, order, or action is useless to it and makes it fail.\n  Allowed: category nouns, colours, materials, clothing, size, type\n  Forbidden: verbs and participles of action (\"turning\", \"walks\", \"is moving\"), temporal words (\"then\", \"starts\", \"after\", \"first\", \"last\"), relative clauses (\"that ...\", \"which ...\", \"who ...\"), ordinals and rank words (\"first\", \"leftmost\", \"nearest\"), and anything over six words.\n  \"the black car that's turning left\" -> \"black car\"\n  \"the cyclist who overtakes the bus\" -> \"cyclist\"\n  \"the two people who walk out of frame on the right\" -> \"person\"\n\nKIND — what makes the expression hard: \"motion\", \"appearance\", \"spatial\", or \"interaction\". Aim for about 35% \"motion\" items: motion is the part a single-frame model cannot already solve, so it is the most valuable, but a list of only motion items is not usable training data either.\n\nEXPECTED_COUNT — how many distinct objects in the clip match the EXPRESSION. Count carefully; this is checked against the detector's answer, and a wrong count discards the item.\n\nIS_NEGATIVE — at most ONE item may be a negative, and only when you also write at least one non-negative item. A negative is an expression that recombines attributes that ARE present in the clip into a whole that is NOT. If the only black car is moving, \"the parked black car\" is a good negative. Its NOUN_PHRASE must still be the category that IS present (\"black car\"), because that is what makes the absence checkable rather than merely unverified. A negative's EXPECTED_COUNT is 0.\n\nRules that discard an item, so follow them exactly:\n- Never write coordinates, box values, pixel positions, or percentages of the frame.\n- Never let a NOUN_PHRASE contain a verb of action, a temporal word, a relative clause, or an ordinal.\n- Never write an EXPRESSION whose object is not visible in the clip, unless it is marked IS_NEGATIVE.\n- Prefer objects that are clearly visible and identifiable over background clutter.\n- Do not describe the camera or the shot; describe objects in the scene.\n","user_prompt":"Watch this clip and write between 1 and 3 items following the rules. Cover different objects rather than restating one object several ways, and include any negative in the same list.\n"},"outputs":{"raw_response":"{\n  \"items\": [\n    {\n      \"expression\": \"the person with long dark hair standing on the street\",\n      \"noun_phrase\": \"person with long dark hair\",\n      \"kind\": \"appearance\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the cafe building illuminated by the streetlamp\",\n      \"noun_phrase\": \"cafe building\",\n      \"kind\": \"appearance\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    }\n  ]\n}","raw_items":[{"expression":"the person with long dark hair standing on the street","noun_phrase":"person with long dark hair","kind":"appearance","expected_count":1,"is_negative":false},{"expression":"the cafe building illuminated by the streetlamp","noun_phrase":"cafe building","kind":"appearance","expected_count":1,"is_negative":false}],"accepted":[{"expression":"the person with long dark hair standing on the street","noun_phrase":"person with long dark hair","kind":"appearance","expected_count":1,"is_negative":false},{"expression":"the cafe building illuminated by the streetlamp","noun_phrase":"cafe building","kind":"appearance","expected_count":1,"is_negative":false}],"gated":[]}},"track":{"inputs":{"tracker":"Sam31MultiplexTracker","model_id":"/mnt/ceph3/ec/jacky/sam3.1","noun_phrases":["cafe building","person with long dark hair"],"expressions_in":2,"frames":49,"sample_fps":4.0,"frame_size":"1920x1080","timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88,10.14,10.4,10.66,10.92,11.18,11.44,11.700000000000001,11.96,12.22,12.48],"session":"one session over 49 JPEG frames; then per phrase: reset_session, add_prompt(text) on frame 0, propagate_in_video","passes":2},"outputs":{"raw_tracks":[{"noun_phrase":"cafe building","object_id":0,"observations":[{"timestamp":7.28,"box":[0.09218750149011612,0.032407406717538834,0.9994792118668556,0.8120369948446751],"occluded":false},{"timestamp":7.54,"box":[0.11562500894069672,0.031481482088565826,0.9994792193174362,0.8268518224358559],"occluded":false},{"timestamp":7.8,"box":[0.1302083432674408,0.03055555559694767,0.999479204416275,0.8435185048729181],"occluded":false},{"timestamp":8.06,"box":[0.06458333879709244,0.03055555559694767,0.9994792118668556,0.8027777764946222],"occluded":false},{"timestamp":8.32,"box":[0.07083334028720856,0.03055555559694767,0.9994792491197586,0.8342592213302851],"occluded":false},{"timestamp":8.58,"box":[0.02239583432674408,0.03055555559694767,0.9994792193174362,0.8046296332031488],"occluded":false},{"timestamp":8.84,"box":[0.3489583432674408,0.03055555559694767,0.999479204416275,0.7703703735023737],"occluded":false},{"timestamp":9.1,"box":[0.38489586114883423,0.03055555559694767,0.9994792342185974,0.8870370481163263],"occluded":false},{"timestamp":9.36,"box":[0.4114583432674408,0.03055555559694767,0.999479204416275,0.7009258959442377],"occluded":false},{"timestamp":9.620000000000001,"box":[0.3739583492279053,0.03055555559694767,0.9994792342185974,0.6990740392357111],"occluded":false},{"timestamp":9.88,"box":[0.30104169249534607,0.03055555559694767,0.999479204416275,0.7583333346992731],"occluded":false},{"timestamp":10.14,"box":[0.21666668355464935,0.03055555559694767,0.9994792193174362,0.7861111257225275],"occluded":false},{"timestamp":10.4,"box":[0.1718750149011612,0.03055555559694767,0.9994792491197586,0.7824074123054743],"occluded":false},{"timestamp":10.66,"box":[0.16614584624767303,0.03055555559694767,0.9994792193174362,0.7564814779907465],"occluded":false},{"timestamp":10.92,"box":[0.1458333432674408,0.03055555559694767,0.999479204416275,0.7861111257225275],"occluded":false}],"scores":[0.7264957427978516,0.7264957427978516,0.7264957427978516,0.7264957427978516,0.7264957427978516,0.7264957427978516,0.7264957427978516,0.7264957427978516,0.7264957427978516,0.7264957427978516,0.7264957427978516,0.7264957427978516,0.7264957427978516,0.7264957427978516,null]},{"noun_phrase":"person with long dark hair","object_id":0,"observations":[{"timestamp":8.84,"box":[0.0,0.12314814329147339,0.16822917759418488,0.9731481075286865],"occluded":false},{"timestamp":9.1,"box":[0.0,0.05000000074505806,0.2880208492279053,0.9777777679264545],"occluded":false},{"timestamp":9.36,"box":[0.0,0.06759259104728699,0.3812500238418579,0.9944444000720978],"occluded":false},{"timestamp":9.620000000000001,"box":[0.0,0.05833333358168602,0.40416669845581055,0.9814814887940884],"occluded":false},{"timestamp":9.88,"box":[0.0,0.029629629105329514,0.3218750059604645,0.9814814738929272],"occluded":false},{"timestamp":10.14,"box":[0.0,0.028703702613711357,0.2718750238418579,0.9749999772757292],"occluded":false},{"timestamp":10.4,"box":[0.0,0.02777777798473835,0.2916666865348816,0.9768518079072237],"occluded":false},{"timestamp":10.66,"box":[0.0,0.03055555559694767,0.2666666805744171,0.9768518302589655],"occluded":false},{"timestamp":10.92,"box":[0.0,0.028703702613711357,0.2526041865348816,0.9759259354323149],"occluded":false},{"timestamp":11.18,"box":[0.0,0.025925925001502037,0.17500001192092896,0.9787036683410406],"occluded":false},{"timestamp":11.44,"box":[0.0,0.17407406866550446,0.1015625074505806,0.9759258776903152],"occluded":false},{"timestamp":11.96,"box":[0.0,0.8305555582046509,0.06927083432674408,0.9768518507480621],"occluded":false},{"timestamp":12.22,"box":[0.0,0.9018518328666687,0.01770833507180214,0.9749999791383743],"occluded":false},{"timestamp":12.48,"box":[0.0,0.654629647731781,0.0937500074505806,0.9814814925193787],"occluded":false}],"scores":[0.8439024686813354,0.8439024686813354,0.8439024686813354,0.8439024686813354,0.8439024686813354,0.8439024686813354,0.8439024686813354,0.8439024686813354,0.8439024686813354,0.8439024686813354,0.8439024686813354,0.8439024686813354,0.8439024686813354,0.8439024686813354]}],"tracklets":[{"noun_phrase":"person with long dark hair","object_id":0,"presentation_id":1,"observations":[{"timestamp":8.84,"box":[0.0,0.12314814329147339,0.16822917759418488,0.9731481075286865],"occluded":false},{"timestamp":9.1,"box":[0.0,0.05000000074505806,0.2880208492279053,0.9777777679264545],"occluded":false},{"timestamp":9.36,"box":[0.0,0.06759259104728699,0.3812500238418579,0.9944444000720978],"occluded":false},{"timestamp":9.620000000000001,"box":[0.0,0.05833333358168602,0.40416669845581055,0.9814814887940884],"occluded":false},{"timestamp":9.88,"box":[0.0,0.029629629105329514,0.3218750059604645,0.9814814738929272],"occluded":false},{"timestamp":10.14,"box":[0.0,0.028703702613711357,0.2718750238418579,0.9749999772757292],"occluded":false},{"timestamp":10.4,"box":[0.0,0.02777777798473835,0.2916666865348816,0.9768518079072237],"occluded":false},{"timestamp":10.66,"box":[0.0,0.03055555559694767,0.2666666805744171,0.9768518302589655],"occluded":false},{"timestamp":10.92,"box":[0.0,0.028703702613711357,0.2526041865348816,0.9759259354323149],"occluded":false},{"timestamp":11.18,"box":[0.0,0.025925925001502037,0.17500001192092896,0.9787036683410406],"occluded":false},{"timestamp":11.44,"box":[0.0,0.17407406866550446,0.1015625074505806,0.9759258776903152],"occluded":false},{"timestamp":11.96,"box":[0.0,0.8305555582046509,0.06927083432674408,0.9768518507480621],"occluded":false},{"timestamp":12.22,"box":[0.0,0.9018518328666687,0.01770833507180214,0.9749999791383743],"occluded":false},{"timestamp":12.48,"box":[0.0,0.654629647731781,0.0937500074505806,0.9814814925193787],"occluded":false}],"mean_score":0.8439024686813354},{"noun_phrase":"cafe building","object_id":0,"presentation_id":2,"observations":[{"timestamp":7.28,"box":[0.09218750149011612,0.032407406717538834,0.9994792118668556,0.8120369948446751],"occluded":false},{"timestamp":7.54,"box":[0.11562500894069672,0.031481482088565826,0.9994792193174362,0.8268518224358559],"occluded":false},{"timestamp":7.8,"box":[0.1302083432674408,0.03055555559694767,0.999479204416275,0.8435185048729181],"occluded":false},{"timestamp":8.06,"box":[0.06458333879709244,0.03055555559694767,0.9994792118668556,0.8027777764946222],"occluded":false},{"timestamp":8.32,"box":[0.07083334028720856,0.03055555559694767,0.9994792491197586,0.8342592213302851],"occluded":false},{"timestamp":8.58,"box":[0.02239583432674408,0.03055555559694767,0.9994792193174362,0.8046296332031488],"occluded":false},{"timestamp":8.84,"box":[0.3489583432674408,0.03055555559694767,0.999479204416275,0.7703703735023737],"occluded":false},{"timestamp":9.1,"box":[0.38489586114883423,0.03055555559694767,0.9994792342185974,0.8870370481163263],"occluded":false},{"timestamp":9.36,"box":[0.4114583432674408,0.03055555559694767,0.999479204416275,0.7009258959442377],"occluded":false},{"timestamp":9.620000000000001,"box":[0.3739583492279053,0.03055555559694767,0.9994792342185974,0.6990740392357111],"occluded":false},{"timestamp":9.88,"box":[0.30104169249534607,0.03055555559694767,0.999479204416275,0.7583333346992731],"occluded":false},{"timestamp":10.14,"box":[0.21666668355464935,0.03055555559694767,0.9994792193174362,0.7861111257225275],"occluded":false},{"timestamp":10.4,"box":[0.1718750149011612,0.03055555559694767,0.9994792491197586,0.7824074123054743],"occluded":false},{"timestamp":10.66,"box":[0.16614584624767303,0.03055555559694767,0.9994792193174362,0.7564814779907465],"occluded":false},{"timestamp":10.92,"box":[0.1458333432674408,0.03055555559694767,0.999479204416275,0.7861111257225275],"occluded":false}],"mean_score":0.7264957427978516}],"dropped":{},"fates":[{"noun_phrase":"cafe building","object_id":0,"frames":15,"mean_score":0.7264957427978516,"mean_area":0.60890608063055,"outcome":"kept","presentation_id":2,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"person with long dark hair","object_id":0,"frames":14,"mean_score":0.8439024686813354,"mean_area":0.1970722153937241,"outcome":"kept","presentation_id":1,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null}],"gates":{"min_track_frames":3,"min_mean_score":0.5,"merge_iou":0.8,"max_tracklets":12}}},"render":{"inputs":{"tracklets":2,"as_video":true,"fps":1.0,"max_frames":30},"outputs":{"media":"/home/zzk/data/video_bbox/vlm_gemini3.5_lite/renders/--cif7Nv6BI_chunk_014_scene_018_choices.mp4","media_bytes":3191073,"frames":[{"index":1,"timestamp":0.0,"file":"--cif7Nv6BI_chunk_014_scene_018_choices.mp4"},{"index":2,"timestamp":1.04,"file":"--cif7Nv6BI_chunk_014_scene_018_choices.mp4"},{"index":3,"timestamp":2.08,"file":"--cif7Nv6BI_chunk_014_scene_018_choices.mp4"},{"index":4,"timestamp":3.12,"file":"--cif7Nv6BI_chunk_014_scene_018_choices.mp4"},{"index":5,"timestamp":4.16,"file":"--cif7Nv6BI_chunk_014_scene_018_choices.mp4"},{"index":6,"timestamp":5.2,"file":"--cif7Nv6BI_chunk_014_scene_018_choices.mp4"},{"index":7,"timestamp":6.24,"file":"--cif7Nv6BI_chunk_014_scene_018_choices.mp4"},{"index":8,"timestamp":7.28,"file":"--cif7Nv6BI_chunk_014_scene_018_choices.mp4"},{"index":9,"timestamp":8.32,"file":"--cif7Nv6BI_chunk_014_scene_018_choices.mp4"},{"index":10,"timestamp":9.36,"file":"--cif7Nv6BI_chunk_014_scene_018_choices.mp4"},{"index":11,"timestamp":10.4,"file":"--cif7Nv6BI_chunk_014_scene_018_choices.mp4"},{"index":12,"timestamp":11.44,"file":"--cif7Nv6BI_chunk_014_scene_018_choices.mp4"},{"index":13,"timestamp":12.48,"file":"--cif7Nv6BI_chunk_014_scene_018_choices.mp4"}]}},"select":{"inputs":{"model":"gemini-3.5-flash-lite","temperature":0.0,"as_video":true,"media_bytes":3191073,"media":["--cif7Nv6BI_chunk_014_scene_018_choices.mp4"],"choice_frames":13,"system_prompt":"You are labelling video data for referring multi-object tracking.\n\nYou will see a numbered sequence of frames from one clip, in time order. Each frame has its own number and timestamp printed in the top-left corner. Candidate objects have been detected for you and drawn on the frames as coloured boxes, each labelled with a number. The same object keeps the same number and the same colour on every frame. Below each frame is a strip of enlarged crops of those same numbered boxes, so you can identify a small or distant object.\n\nFor each description you are given, answer two questions:\n\n1. MATCHING_IDS — which of the numbered objects the description refers to. List every number that matches, or an EMPTY LIST if none of them do. An empty list is a correct and expected answer: some descriptions are deliberately about something that is not in this clip. Do not pick a number just to have picked one, and do not pick a number for an object that merely resembles the description.\n\n2. START_FRAME and END_FRAME — the frame numbers between which the described state actually holds. If the description is \"the car that's turning left\", the interval is the frames in which that car is turning, NOT every frame in which the car is visible. Use the frame numbers printed on the frames. START_FRAME must not be greater than END_FRAME. When MATCHING_IDS is empty, give the full range (the first and last frame numbers); it is ignored.\n\n3. CONFIDENCE — how sure you are of the MATCHING_IDS answer, from 0.0 to 1.0. Be honest: a low number costs the example nothing, and an overconfident wrong answer becomes a wrong label.\n\nRules:\n- Only use numbers that are actually drawn on the frames.\n- Never output coordinates, box values, or pixel positions.\n- Judge motion by comparing the object's position across frames, not by guessing from one frame.\n","user_prompt":"The video has 13 frames, numbered 1 to 13. Each frame is one moment of the clip and has its number burned into the top left corner.\n\nAnswer for each of the following descriptions, in the order given:\n1. the person with long dark hair standing on the street\n2. the cafe building illuminated by the streetlamp","questions":["the person with long dark hair standing on the street","the cafe building illuminated by the streetlamp"]},"outputs":{"raw_response":"{\n  \"answers\": [\n    {\n      \"description_index\": 1,\n      \"matching_ids\": [\n        1\n      ],\n      \"start_frame\": 10,\n      \"end_frame\": 13,\n      \"confidence\": 1.0\n    },\n    {\n      \"description_index\": 2,\n      \"matching_ids\": [\n        2\n      ],\n      \"start_frame\": 8,\n      \"end_frame\": 11,\n      \"confidence\": 1.0\n    }\n  ]\n}","raw_answers":[{"description_index":1,"matching_ids":[1],"start_frame":10,"end_frame":13,"confidence":1.0},{"description_index":2,"matching_ids":[2],"start_frame":8,"end_frame":11,"confidence":1.0}],"answers":[{"expression":"the person with long dark hair standing on the street","matching_presentation_ids":[1],"start_choice_frame":10,"end_choice_frame":13,"confidence":1.0,"self_reported_confidence":1.0},{"expression":"the cafe building illuminated by the streetlamp","matching_presentation_ids":[2],"start_choice_frame":8,"end_choice_frame":11,"confidence":1.0,"self_reported_confidence":1.0}]}},"verify":{"inputs":{"expressions":2,"tracklets":2,"selections":2,"gates":"interval, count, unknown ids, confidence, then geometry (motion)"},"outputs":{"kept":[{"expression":"the person with long dark hair standing on the street","kind":"appearance","is_negative":false,"tracks":[{"timestamps":[9.36,9.620000000000001,9.88,10.14,10.4,10.66,10.92,11.18,11.44,11.96,12.22,12.48],"boxes":[[0.0,0.06759259104728699,0.3812500238418579,0.9944444000720978],[0.0,0.05833333358168602,0.40416669845581055,0.9814814887940884],[0.0,0.029629629105329514,0.3218750059604645,0.9814814738929272],[0.0,0.028703702613711357,0.2718750238418579,0.9749999772757292],[0.0,0.02777777798473835,0.2916666865348816,0.9768518079072237],[0.0,0.03055555559694767,0.2666666805744171,0.9768518302589655],[0.0,0.028703702613711357,0.2526041865348816,0.9759259354323149],[0.0,0.025925925001502037,0.17500001192092896,0.9787036683410406],[0.0,0.17407406866550446,0.1015625074505806,0.9759258776903152],[0.0,0.8305555582046509,0.06927083432674408,0.9768518507480621],[0.0,0.9018518328666687,0.01770833507180214,0.9749999791383743],[0.0,0.654629647731781,0.0937500074505806,0.9814814925193787]]}]},{"expression":"the cafe building illuminated by the streetlamp","kind":"appearance","is_negative":false,"tracks":[{"timestamps":[7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88,10.14,10.4],"boxes":[[0.09218750149011612,0.032407406717538834,0.9994792118668556,0.8120369948446751],[0.11562500894069672,0.031481482088565826,0.9994792193174362,0.8268518224358559],[0.1302083432674408,0.03055555559694767,0.999479204416275,0.8435185048729181],[0.06458333879709244,0.03055555559694767,0.9994792118668556,0.8027777764946222],[0.07083334028720856,0.03055555559694767,0.9994792491197586,0.8342592213302851],[0.02239583432674408,0.03055555559694767,0.9994792193174362,0.8046296332031488],[0.3489583432674408,0.03055555559694767,0.999479204416275,0.7703703735023737],[0.38489586114883423,0.03055555559694767,0.9994792342185974,0.8870370481163263],[0.4114583432674408,0.03055555559694767,0.999479204416275,0.7009258959442377],[0.3739583492279053,0.03055555559694767,0.9994792342185974,0.6990740392357111],[0.30104169249534607,0.03055555559694767,0.999479204416275,0.7583333346992731],[0.21666668355464935,0.03055555559694767,0.9994792193174362,0.7861111257225275],[0.1718750149011612,0.03055555559694767,0.9994792491197586,0.7824074123054743]]}]}],"rejected":{},"rejection_counts":{},"row":{"storage_key":"s3://mv-ytun-external1/silver/test_vehicles/deliveries/videos/delivery1/scene_cutting/v1.0/payload/scenes/--cif7Nv6BI_chunk_014_scene_018.mp4","scene_key":"--cif7Nv6BI_chunk_014_scene_018","category":"test_vehicles","height":1080,"width":1920,"fps":50.0,"num_frames":628,"duration":12.56,"rmot_expressions":[{"expression":"the person with long dark hair standing on the street","kind":"appearance","is_negative":false,"tracks":[{"timestamps":[9.36,9.620000000000001,9.88,10.14,10.4,10.66,10.92,11.18,11.44,11.96,12.22,12.48],"boxes":[[0.0,0.06759259104728699,0.3812500238418579,0.9944444000720978],[0.0,0.05833333358168602,0.40416669845581055,0.9814814887940884],[0.0,0.029629629105329514,0.3218750059604645,0.9814814738929272],[0.0,0.028703702613711357,0.2718750238418579,0.9749999772757292],[0.0,0.02777777798473835,0.2916666865348816,0.9768518079072237],[0.0,0.03055555559694767,0.2666666805744171,0.9768518302589655],[0.0,0.028703702613711357,0.2526041865348816,0.9759259354323149],[0.0,0.025925925001502037,0.17500001192092896,0.9787036683410406],[0.0,0.17407406866550446,0.1015625074505806,0.9759258776903152],[0.0,0.8305555582046509,0.06927083432674408,0.9768518507480621],[0.0,0.9018518328666687,0.01770833507180214,0.9749999791383743],[0.0,0.654629647731781,0.0937500074505806,0.9814814925193787]]}]},{"expression":"the cafe building illuminated by the streetlamp","kind":"appearance","is_negative":false,"tracks":[{"timestamps":[7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88,10.14,10.4],"boxes":[[0.09218750149011612,0.032407406717538834,0.9994792118668556,0.8120369948446751],[0.11562500894069672,0.031481482088565826,0.9994792193174362,0.8268518224358559],[0.1302083432674408,0.03055555559694767,0.999479204416275,0.8435185048729181],[0.06458333879709244,0.03055555559694767,0.9994792118668556,0.8027777764946222],[0.07083334028720856,0.03055555559694767,0.9994792491197586,0.8342592213302851],[0.02239583432674408,0.03055555559694767,0.9994792193174362,0.8046296332031488],[0.3489583432674408,0.03055555559694767,0.999479204416275,0.7703703735023737],[0.38489586114883423,0.03055555559694767,0.9994792342185974,0.8870370481163263],[0.4114583432674408,0.03055555559694767,0.999479204416275,0.7009258959442377],[0.3739583492279053,0.03055555559694767,0.9994792342185974,0.6990740392357111],[0.30104169249534607,0.03055555559694767,0.999479204416275,0.7583333346992731],[0.21666668355464935,0.03055555559694767,0.9994792193174362,0.7861111257225275],[0.1718750149011612,0.03055555559694767,0.9994792491197586,0.7824074123054743]]}]}],"rmot_expression_count":2,"rmot_negative_count":0,"rmot_motion_count":0}}}}}]}