{"run":{"job_id":"86260","host":"GPU-830","model_id":"/mnt/ceph3/ec/jacky/sam3.1","propose_model":"gemini-3.5-flash","select_model":"gemini-3.5-flash","sample_fps":4.0,"choice_frames":8,"clips_requested":20,"output_dir":"/home/zzk/tmp/rmot_smoke_real_86260","wall_seconds":728.14,"measured":{"mean_clip_duration_seconds":17.8658425,"frames_per_clip":69.4,"tracks_per_clip":3.6,"noun_phrase_gate_survival":1.0,"verification_yield":0.576271186440678},"rejections":{"no_candidates_retrieved":14,"count_mismatch":9,"selection_matched":1,"implausible_box":1}},"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","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 dog walking on the cobblestone path\",\n      \"noun_phrase\": \"black dog\",\n      \"kind\": \"motion\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the brown dog standing further up the stone stairs\",\n      \"noun_phrase\": \"brown dog\",\n      \"kind\": \"spatial\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    }\n  ]\n}","raw_items":[{"expression":"the black dog walking on the cobblestone path","noun_phrase":"black dog","kind":"motion","expected_count":1,"is_negative":false},{"expression":"the brown dog standing further up the stone stairs","noun_phrase":"brown dog","kind":"spatial","expected_count":1,"is_negative":false}],"accepted":[{"expression":"the black dog walking on the cobblestone path","noun_phrase":"black dog","kind":"motion","expected_count":1,"is_negative":false},{"expression":"the brown dog standing further up the stone stairs","noun_phrase":"brown dog","kind":"spatial","expected_count":1,"is_negative":false}],"gated":[]}},"track":{"inputs":{"tracker":"Sam31MultiplexTracker","model_id":"/mnt/ceph3/ec/jacky/sam3.1","noun_phrases":["black 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 dog","object_id":0,"observations":[{"timestamp":0.0,"box":[0.3723958432674408,0.4851851761341095,0.6661458611488342,0.8083333075046539],"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":[0.2526041865348816,0.4407407343387604,0.572916716337204,0.852777749300003],"occluded":false},{"timestamp":1.04,"box":[0.21562501788139343,0.45370370149612427,0.5661458671092987,0.8629629611968994],"occluded":false},{"timestamp":1.3,"box":[0.1953125149011612,0.44722220301628113,0.567708358168602,0.8481481075286865],"occluded":false},{"timestamp":1.56,"box":[0.1848958432674408,0.43703702092170715,0.5489583611488342,0.8175925612449646],"occluded":false},{"timestamp":1.82,"box":[0.1666666716337204,0.4120370149612427,0.7614583820104599,0.8324073851108551],"occluded":false},{"timestamp":2.08,"box":[0.13593751192092896,0.4259259104728699,0.5302083790302277,0.9990740418434143],"occluded":false},{"timestamp":2.34,"box":[0.12031250447034836,0.4157407283782959,0.5250000283122063,0.9990740418434143],"occluded":false},{"timestamp":2.6,"box":[0.16354167461395264,0.40833333134651184,0.5614583492279053,0.8148148059844971],"occluded":false},{"timestamp":2.86,"box":[0.24947917461395264,0.3305555582046509,0.9078125357627869,0.7370370328426361],"occluded":false},{"timestamp":3.12,"box":[0.32864585518836975,0.3472222089767456,0.744791716337204,0.7435185015201569],"occluded":false},{"timestamp":3.38,"box":[0.3921875059604645,0.43611109256744385,0.7968750298023224,0.8120370209217072],"occluded":false},{"timestamp":3.64,"box":[0.40677085518836975,0.4851851761341095,0.7708333730697632,0.852777749300003],"occluded":false},{"timestamp":3.9,"box":[0.39531251788139343,0.5175926089286804,0.7072916924953461,0.8666666746139526],"occluded":false},{"timestamp":4.16,"box":[0.39635419845581055,0.5046296119689941,0.6640625596046448,0.8425925672054291],"occluded":false},{"timestamp":4.42,"box":[0.4203125238418579,0.4888888895511627,0.6510417014360428,0.8638888895511627],"occluded":false},{"timestamp":4.68,"box":[0.4348958432674408,0.4749999940395355,0.6354166865348816,0.875],"occluded":false},{"timestamp":4.94,"box":[0.4166666865348816,0.46759259700775146,0.5932291895151138,0.8935185074806213],"occluded":false},{"timestamp":5.2,"box":[0.40000003576278687,0.44814813137054443,0.5354167073965073,0.912962943315506],"occluded":false},{"timestamp":5.46,"box":[0.3348958492279053,0.5055555701255798,0.44739585369825363,0.8666666746139526],"occluded":false},{"timestamp":5.72,"box":[0.27760419249534607,0.5898147821426392,0.3776042014360428,0.9361110627651215],"occluded":false},{"timestamp":5.98,"box":[0.20468750596046448,0.720370352268219,0.3088541775941849,0.9722222089767456],"occluded":false},{"timestamp":6.24,"box":[0.12916667759418488,0.854629635810852,0.2187500149011612,0.9722222238779068],"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]},{"noun_phrase":"black dog","object_id":1,"observations":[{"timestamp":0.52,"box":[0.23541668057441711,0.11203703284263611,0.2812500149011612,0.1666666604578495],"occluded":false},{"timestamp":0.78,"box":[0.28333336114883423,0.10277777910232544,0.32083336263895035,0.1509259268641472],"occluded":false},{"timestamp":1.04,"box":[0.3291666805744171,0.07222221791744232,0.3604166805744171,0.12222221866250038],"occluded":false},{"timestamp":1.3,"box":[0.3812500238418579,0.04722222313284874,0.403645858168602,0.0981481485068798],"occluded":false},{"timestamp":1.56,"box":[0.43697917461395264,0.032407406717538834,0.4614583421498537,0.0694444440305233],"occluded":false}],"scores":[0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116]},{"noun_phrase":"black dog","object_id":2,"observations":[{"timestamp":30.42,"box":[0.510937511920929,0.7990740537643433,0.5546875149011612,0.8666666448116302],"occluded":false},{"timestamp":30.68,"box":[0.5140625238418579,0.8074073791503906,0.55520835891366,0.8824073746800423],"occluded":false},{"timestamp":30.94,"box":[0.5130208730697632,0.8194444179534912,0.5567708760499954,0.8916666358709335],"occluded":false}],"scores":[0.6521739363670349,0.6521739363670349,null]},{"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 dog","object_id":0,"presentation_id":3,"observations":[{"timestamp":0.0,"box":[0.3723958432674408,0.4851851761341095,0.6661458611488342,0.8083333075046539],"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":[0.2526041865348816,0.4407407343387604,0.572916716337204,0.852777749300003],"occluded":false},{"timestamp":1.04,"box":[0.21562501788139343,0.45370370149612427,0.5661458671092987,0.8629629611968994],"occluded":false},{"timestamp":1.3,"box":[0.1953125149011612,0.44722220301628113,0.567708358168602,0.8481481075286865],"occluded":false},{"timestamp":1.56,"box":[0.1848958432674408,0.43703702092170715,0.5489583611488342,0.8175925612449646],"occluded":false},{"timestamp":1.82,"box":[0.1666666716337204,0.4120370149612427,0.7614583820104599,0.8324073851108551],"occluded":false},{"timestamp":2.08,"box":[0.13593751192092896,0.4259259104728699,0.5302083790302277,0.9990740418434143],"occluded":false},{"timestamp":2.34,"box":[0.12031250447034836,0.4157407283782959,0.5250000283122063,0.9990740418434143],"occluded":false},{"timestamp":2.6,"box":[0.16354167461395264,0.40833333134651184,0.5614583492279053,0.8148148059844971],"occluded":false},{"timestamp":2.86,"box":[0.24947917461395264,0.3305555582046509,0.9078125357627869,0.7370370328426361],"occluded":false},{"timestamp":3.12,"box":[0.32864585518836975,0.3472222089767456,0.744791716337204,0.7435185015201569],"occluded":false},{"timestamp":3.38,"box":[0.3921875059604645,0.43611109256744385,0.7968750298023224,0.8120370209217072],"occluded":false},{"timestamp":3.64,"box":[0.40677085518836975,0.4851851761341095,0.7708333730697632,0.852777749300003],"occluded":false},{"timestamp":3.9,"box":[0.39531251788139343,0.5175926089286804,0.7072916924953461,0.8666666746139526],"occluded":false},{"timestamp":4.16,"box":[0.39635419845581055,0.5046296119689941,0.6640625596046448,0.8425925672054291],"occluded":false},{"timestamp":4.42,"box":[0.4203125238418579,0.4888888895511627,0.6510417014360428,0.8638888895511627],"occluded":false},{"timestamp":4.68,"box":[0.4348958432674408,0.4749999940395355,0.6354166865348816,0.875],"occluded":false},{"timestamp":4.94,"box":[0.4166666865348816,0.46759259700775146,0.5932291895151138,0.8935185074806213],"occluded":false},{"timestamp":5.2,"box":[0.40000003576278687,0.44814813137054443,0.5354167073965073,0.912962943315506],"occluded":false},{"timestamp":5.46,"box":[0.3348958492279053,0.5055555701255798,0.44739585369825363,0.8666666746139526],"occluded":false},{"timestamp":5.72,"box":[0.27760419249534607,0.5898147821426392,0.3776042014360428,0.9361110627651215],"occluded":false},{"timestamp":5.98,"box":[0.20468750596046448,0.720370352268219,0.3088541775941849,0.9722222089767456],"occluded":false},{"timestamp":6.24,"box":[0.12916667759418488,0.854629635810852,0.2187500149011612,0.9722222238779068],"occluded":false}],"mean_score":0.9609756469726562}],"dropped":{"merged_duplicate":2},"fates":[{"noun_phrase":"black dog","object_id":0,"frames":25,"mean_score":0.9609756469726562,"mean_area":0.1234315630525993,"outcome":"kept","presentation_id":3,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"black dog","object_id":1,"frames":5,"mean_score":0.8904109597206116,"mean_area":0.0015838156103524432,"outcome":"merged_duplicate","presentation_id":null,"merged_into_phrase":"brown dog","merged_into_object_id":0,"merge_iou":0.9353130326577878},{"noun_phrase":"black dog","object_id":2,"frames":3,"mean_score":0.6521739363670349,"mean_area":0.003067611918445484,"outcome":"merged_duplicate","presentation_id":null,"merged_into_phrase":"brown dog","merged_into_object_id":1,"merge_iou":0.9707962609241076},{"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/tmp/rmot_smoke_real_86260/renders/--cif7Nv6BI_chunk_003_scene_023_choices.mp4","media_bytes":11580614,"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","temperature":0.0,"as_video":true,"media_bytes":11580614,"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 dog walking on the cobblestone path\n2. the brown dog standing further up the stone stairs","questions":["the black dog walking on the cobblestone path","the brown dog standing further up the stone stairs"]},"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\": 2,\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":2,"end_frame":17,"confidence":1.0}],"answers":[{"expression":"the black dog walking on the cobblestone path","matching_presentation_ids":[3],"start_choice_frame":1,"end_choice_frame":6,"confidence":1.0,"self_reported_confidence":1.0},{"expression":"the brown dog standing further up the stone stairs","matching_presentation_ids":[1],"start_choice_frame":2,"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 dog walking on the cobblestone path","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":[[0.3723958432674408,0.4851851761341095,0.6661458611488342,0.8083333075046539],[0.33385419845581055,0.470370352268219,0.6234375536441803,0.8083333075046539],[0.2953125238418579,0.45185184478759766,0.5906250476837158,0.8620370328426361],[0.2526041865348816,0.4407407343387604,0.572916716337204,0.852777749300003],[0.21562501788139343,0.45370370149612427,0.5661458671092987,0.8629629611968994],[0.1953125149011612,0.44722220301628113,0.567708358168602,0.8481481075286865],[0.1848958432674408,0.43703702092170715,0.5489583611488342,0.8175925612449646],[0.1666666716337204,0.4120370149612427,0.7614583820104599,0.8324073851108551],[0.13593751192092896,0.4259259104728699,0.5302083790302277,0.9990740418434143],[0.12031250447034836,0.4157407283782959,0.5250000283122063,0.9990740418434143],[0.16354167461395264,0.40833333134651184,0.5614583492279053,0.8148148059844971],[0.24947917461395264,0.3305555582046509,0.9078125357627869,0.7370370328426361],[0.32864585518836975,0.3472222089767456,0.744791716337204,0.7435185015201569],[0.3921875059604645,0.43611109256744385,0.7968750298023224,0.8120370209217072],[0.40677085518836975,0.4851851761341095,0.7708333730697632,0.852777749300003],[0.39531251788139343,0.5175926089286804,0.7072916924953461,0.8666666746139526],[0.39635419845581055,0.5046296119689941,0.6640625596046448,0.8425925672054291],[0.4203125238418579,0.4888888895511627,0.6510417014360428,0.8638888895511627],[0.4348958432674408,0.4749999940395355,0.6354166865348816,0.875],[0.4166666865348816,0.46759259700775146,0.5932291895151138,0.8935185074806213],[0.40000003576278687,0.44814813137054443,0.5354167073965073,0.912962943315506],[0.3348958492279053,0.5055555701255798,0.44739585369825363,0.8666666746139526]]}]},{"expression":"the brown dog standing further up the stone stairs","kind":"spatial","is_negative":false,"tracks":[{"timestamps":[1.04,1.3,1.56,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88,10.14,10.4,10.66,10.92,11.18,11.44,11.700000000000001,11.96,12.22,12.48,12.74,13.0,13.26,13.52,13.780000000000001,14.040000000000001,14.3,14.56,14.82,15.08,15.34,15.6,15.860000000000001,16.12,16.38,16.64,16.9,17.16,17.42,17.68],"boxes":[[0.3296875059604645,0.07222221791744232,0.35989584028720856,0.12129629030823708],[0.3812500238418579,0.04814814776182175,0.40312502533197403,0.0981481485068798],[0.43906253576278687,0.032407406717538834,0.4598958697170019,0.06851851567626],[0.7026041746139526,0.031481482088565826,0.7723958417773247,0.06481481343507767],[0.6427083611488342,0.14444443583488464,0.7026041969656944,0.19074073061347008],[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 dog walking on the cobblestone path","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":[[0.3723958432674408,0.4851851761341095,0.6661458611488342,0.8083333075046539],[0.33385419845581055,0.470370352268219,0.6234375536441803,0.8083333075046539],[0.2953125238418579,0.45185184478759766,0.5906250476837158,0.8620370328426361],[0.2526041865348816,0.4407407343387604,0.572916716337204,0.852777749300003],[0.21562501788139343,0.45370370149612427,0.5661458671092987,0.8629629611968994],[0.1953125149011612,0.44722220301628113,0.567708358168602,0.8481481075286865],[0.1848958432674408,0.43703702092170715,0.5489583611488342,0.8175925612449646],[0.1666666716337204,0.4120370149612427,0.7614583820104599,0.8324073851108551],[0.13593751192092896,0.4259259104728699,0.5302083790302277,0.9990740418434143],[0.12031250447034836,0.4157407283782959,0.5250000283122063,0.9990740418434143],[0.16354167461395264,0.40833333134651184,0.5614583492279053,0.8148148059844971],[0.24947917461395264,0.3305555582046509,0.9078125357627869,0.7370370328426361],[0.32864585518836975,0.3472222089767456,0.744791716337204,0.7435185015201569],[0.3921875059604645,0.43611109256744385,0.7968750298023224,0.8120370209217072],[0.40677085518836975,0.4851851761341095,0.7708333730697632,0.852777749300003],[0.39531251788139343,0.5175926089286804,0.7072916924953461,0.8666666746139526],[0.39635419845581055,0.5046296119689941,0.6640625596046448,0.8425925672054291],[0.4203125238418579,0.4888888895511627,0.6510417014360428,0.8638888895511627],[0.4348958432674408,0.4749999940395355,0.6354166865348816,0.875],[0.4166666865348816,0.46759259700775146,0.5932291895151138,0.8935185074806213],[0.40000003576278687,0.44814813137054443,0.5354167073965073,0.912962943315506],[0.3348958492279053,0.5055555701255798,0.44739585369825363,0.8666666746139526]]}]},{"expression":"the brown dog standing further up the stone stairs","kind":"spatial","is_negative":false,"tracks":[{"timestamps":[1.04,1.3,1.56,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88,10.14,10.4,10.66,10.92,11.18,11.44,11.700000000000001,11.96,12.22,12.48,12.74,13.0,13.26,13.52,13.780000000000001,14.040000000000001,14.3,14.56,14.82,15.08,15.34,15.6,15.860000000000001,16.12,16.38,16.64,16.9,17.16,17.42,17.68],"boxes":[[0.3296875059604645,0.07222221791744232,0.35989584028720856,0.12129629030823708],[0.3812500238418579,0.04814814776182175,0.40312502533197403,0.0981481485068798],[0.43906253576278687,0.032407406717538834,0.4598958697170019,0.06851851567626],[0.7026041746139526,0.031481482088565826,0.7723958417773247,0.06481481343507767],[0.6427083611488342,0.14444443583488464,0.7026041969656944,0.19074073061347008],[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":1}}}}},{"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","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 hand that appears in front of the camera at the start\",\n      \"noun_phrase\": \"hand\",\n      \"kind\": \"motion\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the white car parked on the street\",\n      \"noun_phrase\": \"white car\",\n      \"kind\": \"spatial\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the red car driving down the street\",\n      \"noun_phrase\": \"red car\",\n      \"kind\": \"motion\",\n      \"expected_count\": 0,\n      \"is_negative\": true\n    }\n  ]\n}","raw_items":[{"expression":"the hand that appears in front of the camera at the start","noun_phrase":"hand","kind":"motion","expected_count":1,"is_negative":false},{"expression":"the white car parked on the street","noun_phrase":"white car","kind":"spatial","expected_count":1,"is_negative":false},{"expression":"the red car driving down the street","noun_phrase":"red car","kind":"motion","expected_count":0,"is_negative":true}],"accepted":[{"expression":"the hand that appears in front of the camera at the start","noun_phrase":"hand","kind":"motion","expected_count":1,"is_negative":false},{"expression":"the white car parked on the street","noun_phrase":"white car","kind":"spatial","expected_count":1,"is_negative":false},{"expression":"the red car driving down the street","noun_phrase":"red car","kind":"motion","expected_count":0,"is_negative":true}],"gated":[]}},"track":{"inputs":{"tracker":"Sam31MultiplexTracker","model_id":"/mnt/ceph3/ec/jacky/sam3.1","noun_phrases":["hand","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":"hand","object_id":0,"observations":[{"timestamp":0.0,"box":[0.24635417759418488,0.5138888955116272,0.5750000327825546,0.9703703820705414],"occluded":false},{"timestamp":0.26,"box":[0.32083335518836975,0.5222222208976746,0.6411458849906921,0.9712962806224823],"occluded":false},{"timestamp":0.52,"box":[0.2708333432674408,0.5249999761581421,0.573437511920929,0.970370352268219],"occluded":false},{"timestamp":0.78,"box":[0.17239584028720856,0.54537034034729,0.473958358168602,0.9703703224658966],"occluded":false},{"timestamp":1.04,"box":[0.10833334177732468,0.614814817905426,0.408854179084301,0.9990740716457367],"occluded":false}],"scores":[0.9298245906829834,0.9298245906829834,0.9298245906829834,0.9298245906829834,0.9298245906829834]},{"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":"white car","object_id":0,"presentation_id":1,"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":"red car","object_id":0,"presentation_id":2,"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},{"noun_phrase":"white car","object_id":1,"presentation_id":3,"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":"hand","object_id":0,"presentation_id":4,"observations":[{"timestamp":0.0,"box":[0.24635417759418488,0.5138888955116272,0.5750000327825546,0.9703703820705414],"occluded":false},{"timestamp":0.26,"box":[0.32083335518836975,0.5222222208976746,0.6411458849906921,0.9712962806224823],"occluded":false},{"timestamp":0.52,"box":[0.2708333432674408,0.5249999761581421,0.573437511920929,0.970370352268219],"occluded":false},{"timestamp":0.78,"box":[0.17239584028720856,0.54537034034729,0.473958358168602,0.9703703224658966],"occluded":false},{"timestamp":1.04,"box":[0.10833334177732468,0.614814817905426,0.408854179084301,0.9990740716457367],"occluded":false}],"mean_score":0.9298245906829834}],"dropped":{},"fates":[{"noun_phrase":"hand","object_id":0,"frames":5,"mean_score":0.9298245906829834,"mean_area":0.13445554129135448,"outcome":"kept","presentation_id":4,"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":2,"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":1,"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":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":4,"as_video":true,"fps":1.0,"max_frames":30},"outputs":{"media":"/home/zzk/tmp/rmot_smoke_real_86260/renders/--cif7Nv6BI_chunk_000_scene_044_choices.mp4","media_bytes":3190498,"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","temperature":0.0,"as_video":true,"media_bytes":3190498,"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 hand that appears in front of the camera at the start\n2. the white car parked on the street\n3. the red car driving down the street","questions":["the hand that appears in front of the camera at the start","the white car parked on the street","the red car driving down the street"]},"outputs":{"raw_response":"{\n  \"answers\": [\n    {\n      \"description_index\": 1,\n      \"matching_ids\": [\n        4\n      ],\n      \"start_frame\": 1,\n      \"end_frame\": 2,\n      \"confidence\": 1.0\n    },\n    {\n      \"description_index\": 2,\n      \"matching_ids\": [\n        1\n      ],\n      \"start_frame\": 6,\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":[4],"start_frame":1,"end_frame":2,"confidence":1.0},{"description_index":2,"matching_ids":[1],"start_frame":6,"end_frame":11,"confidence":1.0},{"description_index":3,"matching_ids":[],"start_frame":1,"end_frame":11,"confidence":1.0}],"answers":[{"expression":"the hand that appears in front of the camera at the start","matching_presentation_ids":[4],"start_choice_frame":1,"end_choice_frame":2,"confidence":1.0,"self_reported_confidence":1.0},{"expression":"the white car parked on the street","matching_presentation_ids":[1],"start_choice_frame":6,"end_choice_frame":11,"confidence":1.0,"self_reported_confidence":1.0},{"expression":"the red car driving down 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":4,"selections":3,"gates":"interval, count, unknown ids, confidence, then geometry (motion)"},"outputs":{"kept":[{"expression":"the hand that appears in front of the camera at the start","kind":"motion","is_negative":false,"tracks":[{"timestamps":[0.0,0.26,0.52,0.78,1.04],"boxes":[[0.24635417759418488,0.5138888955116272,0.5750000327825546,0.9703703820705414],[0.32083335518836975,0.5222222208976746,0.6411458849906921,0.9712962806224823],[0.2708333432674408,0.5249999761581421,0.573437511920929,0.970370352268219],[0.17239584028720856,0.54537034034729,0.473958358168602,0.9703703224658966],[0.10833334177732468,0.614814817905426,0.408854179084301,0.9990740716457367]]}]},{"expression":"the white car parked on the street","kind":"spatial","is_negative":false,"tracks":[{"timestamps":[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":[[0.22760418057441711,0.520370364189148,0.39895835518836975,0.6574074029922485],[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 red car driving down the street","kind":"motion","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 hand that appears in front of the camera at the start","kind":"motion","is_negative":false,"tracks":[{"timestamps":[0.0,0.26,0.52,0.78,1.04],"boxes":[[0.24635417759418488,0.5138888955116272,0.5750000327825546,0.9703703820705414],[0.32083335518836975,0.5222222208976746,0.6411458849906921,0.9712962806224823],[0.2708333432674408,0.5249999761581421,0.573437511920929,0.970370352268219],[0.17239584028720856,0.54537034034729,0.473958358168602,0.9703703224658966],[0.10833334177732468,0.614814817905426,0.408854179084301,0.9990740716457367]]}]},{"expression":"the white car parked on the street","kind":"spatial","is_negative":false,"tracks":[{"timestamps":[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":[[0.22760418057441711,0.520370364189148,0.39895835518836975,0.6574074029922485],[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 red car driving down the street","kind":"motion","is_negative":true,"tracks":[]}],"rmot_expression_count":3,"rmot_negative_count":1,"rmot_motion_count":2}}}}},{"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","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 on the right side of the path\",\n      \"noun_phrase\": \"stone wall\",\n      \"kind\": \"spatial\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the collapsed wooden roof above\",\n      \"noun_phrase\": \"wooden roof\",\n      \"kind\": \"appearance\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the person walking ahead of the camera\",\n      \"noun_phrase\": \"person\",\n      \"kind\": \"appearance\",\n      \"expected_count\": 0,\n      \"is_negative\": true\n    }\n  ]\n}","raw_items":[{"expression":"the stone wall on the right side of the path","noun_phrase":"stone wall","kind":"spatial","expected_count":1,"is_negative":false},{"expression":"the collapsed wooden roof above","noun_phrase":"wooden roof","kind":"appearance","expected_count":1,"is_negative":false},{"expression":"the person walking ahead of the camera","noun_phrase":"person","kind":"appearance","expected_count":0,"is_negative":true}],"accepted":[{"expression":"the stone wall on the right side of the path","noun_phrase":"stone wall","kind":"spatial","expected_count":1,"is_negative":false},{"expression":"the collapsed wooden roof above","noun_phrase":"wooden roof","kind":"appearance","expected_count":1,"is_negative":false},{"expression":"the person walking ahead of the camera","noun_phrase":"person","kind":"appearance","expected_count":0,"is_negative":true}],"gated":[]}},"track":{"inputs":{"tracker":"Sam31MultiplexTracker","model_id":"/mnt/ceph3/ec/jacky/sam3.1","noun_phrases":["person","stone wall","wooden roof"],"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]}],"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":1,"presentation_id":2,"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":"stone wall","object_id":2,"presentation_id":3,"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}],"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":2,"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":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/tmp/rmot_smoke_real_86260/renders/--cif7Nv6BI_chunk_014_scene_002_choices.mp4","media_bytes":4134404,"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","temperature":0.0,"as_video":true,"media_bytes":4134404,"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 on the right side of the path\n2. the collapsed wooden roof above\n3. the person walking ahead of the camera","questions":["the stone wall on the right side of the path","the collapsed wooden roof above","the person walking ahead of the camera"]},"outputs":{"raw_response":"{\n  \"answers\": [\n    {\n      \"description_index\": 1,\n      \"matching_ids\": [\n        1\n      ],\n      \"start_frame\": 1,\n      \"end_frame\": 7,\n      \"confidence\": 1.0\n    },\n    {\n      \"description_index\": 2,\n      \"matching_ids\": [],\n      \"start_frame\": 1,\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":7,"confidence":1.0},{"description_index":2,"matching_ids":[],"start_frame":1,"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 on the right side of the path","matching_presentation_ids":[1],"start_choice_frame":1,"end_choice_frame":7,"confidence":1.0,"self_reported_confidence":1.0},{"expression":"the collapsed wooden roof above","matching_presentation_ids":[],"start_choice_frame":1,"end_choice_frame":16,"confidence":0.9,"self_reported_confidence":0.9},{"expression":"the person walking ahead of the camera","matching_presentation_ids":[],"start_choice_frame":1,"end_choice_frame":16,"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 stone wall on the right side of the path","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],"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]]}]}],"rejected":{"the person walking ahead of the camera":"no_candidates_retrieved","the collapsed wooden roof above":"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 on the right side of the path","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],"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]]}]}],"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","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 bearded man standing on the stone path\",\n      \"noun_phrase\": \"bearded man\",\n      \"kind\": \"appearance\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the dark dog walking up the path\",\n      \"noun_phrase\": \"dark dog\",\n      \"kind\": \"motion\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the white dog on the path\",\n      \"noun_phrase\": \"white dog\",\n      \"kind\": \"appearance\",\n      \"expected_count\": 0,\n      \"is_negative\": true\n    }\n  ]\n}","raw_items":[{"expression":"the bearded man standing on the stone path","noun_phrase":"bearded man","kind":"appearance","expected_count":1,"is_negative":false},{"expression":"the dark dog walking up the path","noun_phrase":"dark dog","kind":"motion","expected_count":1,"is_negative":false},{"expression":"the white dog on the path","noun_phrase":"white dog","kind":"appearance","expected_count":0,"is_negative":true}],"accepted":[{"expression":"the bearded man standing on the stone path","noun_phrase":"bearded man","kind":"appearance","expected_count":1,"is_negative":false},{"expression":"the dark dog walking up the path","noun_phrase":"dark dog","kind":"motion","expected_count":1,"is_negative":false},{"expression":"the white dog on the path","noun_phrase":"white dog","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","dark dog","white dog"],"expressions_in":3,"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":3},"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]}],"tracklets":[{"noun_phrase":"dark dog","object_id":0,"presentation_id":1,"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}],"mean_score":0.7852349281311035}],"dropped":{},"fates":[{"noun_phrase":"dark dog","object_id":0,"frames":21,"mean_score":0.7852349281311035,"mean_area":0.009223986185546727,"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/tmp/rmot_smoke_real_86260/renders/--cif7Nv6BI_chunk_008_scene_014_choices.mp4","media_bytes":3580385,"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","temperature":0.0,"as_video":true,"media_bytes":3580385,"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 bearded man standing on the stone path\n2. the dark dog walking up the path\n3. the white dog on the path","questions":["the bearded man standing on the stone path","the dark dog walking up the path","the white dog on the path"]},"outputs":{"raw_response":"{\n  \"answers\": [\n    {\n      \"description_index\": 1,\n      \"matching_ids\": [],\n      \"start_frame\": 1,\n      \"end_frame\": 13,\n      \"confidence\": 1.0\n    },\n    {\n      \"description_index\": 2,\n      \"matching_ids\": [\n        1\n      ],\n      \"start_frame\": 4,\n      \"end_frame\": 13,\n      \"confidence\": 1.0\n    },\n    {\n      \"description_index\": 3,\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":[],"start_frame":1,"end_frame":13,"confidence":1.0},{"description_index":2,"matching_ids":[1],"start_frame":4,"end_frame":13,"confidence":1.0},{"description_index":3,"matching_ids":[],"start_frame":1,"end_frame":13,"confidence":1.0}],"answers":[{"expression":"the bearded man standing on the stone path","matching_presentation_ids":[],"start_choice_frame":1,"end_choice_frame":13,"confidence":1.0,"self_reported_confidence":1.0},{"expression":"the dark dog walking up the path","matching_presentation_ids":[1],"start_choice_frame":4,"end_choice_frame":13,"confidence":1.0,"self_reported_confidence":1.0},{"expression":"the white dog on the path","matching_presentation_ids":[],"start_choice_frame":1,"end_choice_frame":13,"confidence":1.0,"self_reported_confidence":1.0}]}},"verify":{"inputs":{"expressions":3,"tracklets":1,"selections":3,"gates":"interval, count, unknown ids, confidence, then geometry (motion)"},"outputs":{"kept":[{"expression":"the dark dog walking up the path","kind":"motion","is_negative":false,"tracks":[{"timestamps":[2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,9.620000000000001,10.4,10.66,10.92,11.18,11.44,11.700000000000001],"boxes":[[0.5057291984558105,0.5185185074806213,0.5213541984558105,0.574999988079071],[0.4828125238418579,0.5,0.498958358541131,0.5537037029862404],[0.4869791865348816,0.4851851761341095,0.5036458540707827,0.5435185097157955],[0.49010419845581055,0.48055553436279297,0.5062500331550837,0.5462962761521339],[0.4911458492279053,0.4888888895511627,0.5098958499729633,0.5500000007450581],[0.47864586114883423,0.4879629611968994,0.5020833630114794,0.5537037029862404],[0.4713541865348816,0.49444442987442017,0.5072916895151138,0.5583333149552345],[0.46302086114883423,0.5027777552604675,0.5098958648741245,0.5675925686955452],[0.4557291865348816,0.5342592597007751,0.5088541880249977,0.6009259223937988],[0.46302086114883423,0.5583333373069763,0.5151041969656944,0.6240740790963173],[0.4723958671092987,0.5509259104728699,0.527083370834589,0.6157407239079475],[0.4661458432674408,0.54537034034729,0.5203125141561031,0.6101851537823677],[0.483854204416275,0.5879629254341125,0.5338542088866234,0.6555555164813995],[0.659375011920929,0.7620370388031006,0.7312500178813934,0.8509259223937988],[0.5776041746139526,0.3759259283542633,0.8171875178813934,0.9749999940395355],[0.6390625238418579,0.5407407283782959,0.6817708611488342,0.5990740619599819],[0.5213541984558105,0.5222222208976746,0.5671875327825546,0.5787037014961243],[0.49531251192092896,0.5175926089286804,0.5390625149011612,0.5731481648981571],[0.4791666865348816,0.550000011920929,0.5223958566784859,0.6064814925193787],[0.47864586114883423,0.550000011920929,0.5208333618938923,0.6055555678904057],[0.476041704416275,0.550000011920929,0.5171875394880772,0.6055555678904057]]}]}],"rejected":{"the white dog on the path":"no_candidates_retrieved","the bearded man standing on the stone path":"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_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 dark dog walking up the path","kind":"motion","is_negative":false,"tracks":[{"timestamps":[2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,9.620000000000001,10.4,10.66,10.92,11.18,11.44,11.700000000000001],"boxes":[[0.5057291984558105,0.5185185074806213,0.5213541984558105,0.574999988079071],[0.4828125238418579,0.5,0.498958358541131,0.5537037029862404],[0.4869791865348816,0.4851851761341095,0.5036458540707827,0.5435185097157955],[0.49010419845581055,0.48055553436279297,0.5062500331550837,0.5462962761521339],[0.4911458492279053,0.4888888895511627,0.5098958499729633,0.5500000007450581],[0.47864586114883423,0.4879629611968994,0.5020833630114794,0.5537037029862404],[0.4713541865348816,0.49444442987442017,0.5072916895151138,0.5583333149552345],[0.46302086114883423,0.5027777552604675,0.5098958648741245,0.5675925686955452],[0.4557291865348816,0.5342592597007751,0.5088541880249977,0.6009259223937988],[0.46302086114883423,0.5583333373069763,0.5151041969656944,0.6240740790963173],[0.4723958671092987,0.5509259104728699,0.527083370834589,0.6157407239079475],[0.4661458432674408,0.54537034034729,0.5203125141561031,0.6101851537823677],[0.483854204416275,0.5879629254341125,0.5338542088866234,0.6555555164813995],[0.659375011920929,0.7620370388031006,0.7312500178813934,0.8509259223937988],[0.5776041746139526,0.3759259283542633,0.8171875178813934,0.9749999940395355],[0.6390625238418579,0.5407407283782959,0.6817708611488342,0.5990740619599819],[0.5213541984558105,0.5222222208976746,0.5671875327825546,0.5787037014961243],[0.49531251192092896,0.5175926089286804,0.5390625149011612,0.5731481648981571],[0.4791666865348816,0.550000011920929,0.5223958566784859,0.6064814925193787],[0.47864586114883423,0.550000011920929,0.5208333618938923,0.6055555678904057],[0.476041704416275,0.550000011920929,0.5171875394880772,0.6055555678904057]]}]}],"rmot_expression_count":1,"rmot_negative_count":0,"rmot_motion_count":1}}}}},{"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","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 silver rooster ornament on top of the church roof\",\n      \"noun_phrase\": \"silver rooster statue\",\n      \"kind\": \"spatial\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the white truck parked on the right side of the street\",\n      \"noun_phrase\": \"white truck\",\n      \"kind\": \"spatial\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"a person walking down the stone street\",\n      \"noun_phrase\": \"person\",\n      \"kind\": \"motion\",\n      \"expected_count\": 0,\n      \"is_negative\": true\n    }\n  ]\n}","raw_items":[{"expression":"the silver rooster ornament on top of the church roof","noun_phrase":"silver rooster statue","kind":"spatial","expected_count":1,"is_negative":false},{"expression":"the white truck parked on the right side of the street","noun_phrase":"white truck","kind":"spatial","expected_count":1,"is_negative":false},{"expression":"a person walking down the stone street","noun_phrase":"person","kind":"motion","expected_count":0,"is_negative":true}],"accepted":[{"expression":"the silver rooster ornament on top of the church roof","noun_phrase":"silver rooster statue","kind":"spatial","expected_count":1,"is_negative":false},{"expression":"the white truck parked on the right side of the street","noun_phrase":"white truck","kind":"spatial","expected_count":1,"is_negative":false},{"expression":"a person walking down the stone street","noun_phrase":"person","kind":"motion","expected_count":0,"is_negative":true}],"gated":[]}},"track":{"inputs":{"tracker":"Sam31MultiplexTracker","model_id":"/mnt/ceph3/ec/jacky/sam3.1","noun_phrases":["person","silver rooster statue","white truck"],"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":"silver rooster statue","object_id":0,"observations":[{"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":[0.49635419249534607,0.4731481373310089,0.6890625357627869,0.7129629552364349],"occluded":false}],"scores":[0.7887788414955139,0.7887788414955139,null]},{"noun_phrase":"silver rooster statue","object_id":1,"observations":[{"timestamp":7.28,"box":[0.5041667222976685,0.4555555582046509,0.6890625655651093,0.7129629552364349],"occluded":false},{"timestamp":7.54,"box":[0.5010417103767395,0.4564814865589142,0.5203125439584255,0.5166666693985462],"occluded":false},{"timestamp":7.8,"box":[0.499479204416275,0.4574073851108551,0.518750037997961,0.5175925679504871],"occluded":false}],"scores":[0.6717948913574219,0.6717948913574219,null]},{"noun_phrase":"silver rooster statue","object_id":2,"observations":[{"timestamp":11.44,"box":[0.5609375238418579,0.3712962865829468,0.5713541908189654,0.3990740645676851],"occluded":false},{"timestamp":11.700000000000001,"box":[0.5750000476837158,0.3722222149372101,0.5859375484287739,0.3972222153097391],"occluded":false},{"timestamp":11.96,"box":[0.5791667103767395,0.32685184478759766,0.5901042111217976,0.3462962880730629],"occluded":false}],"scores":[0.6800000071525574,0.6800000071525574,null]},{"noun_phrase":"white truck","object_id":0,"observations":[{"timestamp":12.22,"box":[0.8562500476837158,0.5981481671333313,0.9994792193174362,0.9703703820705414],"occluded":false},{"timestamp":12.48,"box":[0.8375000357627869,0.5472221970558167,0.999479204416275,0.9712962508201599],"occluded":false},{"timestamp":12.74,"box":[0.8776041865348816,0.5092592239379883,0.9994791895151138,0.9722221791744232],"occluded":false},{"timestamp":13.0,"box":[0.9166666865348816,0.5703703761100769,0.9994791895151138,0.9722222089767456],"occluded":false},{"timestamp":13.26,"box":[0.9458333849906921,0.6314814686775208,0.9994792193174362,0.9722222089767456],"occluded":false},{"timestamp":13.52,"box":[0.9447917342185974,0.6388888955116272,0.9994792379438877,0.9722222089767456],"occluded":false},{"timestamp":13.780000000000001,"box":[0.9369792342185974,0.6564814448356628,0.9994792342185974,0.9990740418434143],"occluded":false}],"scores":[0.7887788414955139,0.7887788414955139,0.7887788414955139,0.7887788414955139,0.7887788414955139,0.7887788414955139,0.7887788414955139]}],"tracklets":[{"noun_phrase":"silver rooster statue","object_id":2,"presentation_id":1,"observations":[{"timestamp":11.44,"box":[0.5609375238418579,0.3712962865829468,0.5713541908189654,0.3990740645676851],"occluded":false},{"timestamp":11.700000000000001,"box":[0.5750000476837158,0.3722222149372101,0.5859375484287739,0.3972222153097391],"occluded":false},{"timestamp":11.96,"box":[0.5791667103767395,0.32685184478759766,0.5901042111217976,0.3462962880730629],"occluded":false}],"mean_score":0.6800000071525574},{"noun_phrase":"silver rooster statue","object_id":1,"presentation_id":2,"observations":[{"timestamp":7.28,"box":[0.5041667222976685,0.4555555582046509,0.6890625655651093,0.7129629552364349],"occluded":false},{"timestamp":7.54,"box":[0.5010417103767395,0.4564814865589142,0.5203125439584255,0.5166666693985462],"occluded":false},{"timestamp":7.8,"box":[0.499479204416275,0.4574073851108551,0.518750037997961,0.5175925679504871],"occluded":false}],"mean_score":0.6717948913574219},{"noun_phrase":"white truck","object_id":0,"presentation_id":3,"observations":[{"timestamp":12.22,"box":[0.8562500476837158,0.5981481671333313,0.9994792193174362,0.9703703820705414],"occluded":false},{"timestamp":12.48,"box":[0.8375000357627869,0.5472221970558167,0.999479204416275,0.9712962508201599],"occluded":false},{"timestamp":12.74,"box":[0.8776041865348816,0.5092592239379883,0.9994791895151138,0.9722221791744232],"occluded":false},{"timestamp":13.0,"box":[0.9166666865348816,0.5703703761100769,0.9994791895151138,0.9722222089767456],"occluded":false},{"timestamp":13.26,"box":[0.9458333849906921,0.6314814686775208,0.9994792193174362,0.9722222089767456],"occluded":false},{"timestamp":13.52,"box":[0.9447917342185974,0.6388888955116272,0.9994792379438877,0.9722222089767456],"occluded":false},{"timestamp":13.780000000000001,"box":[0.9369792342185974,0.6564814448356628,0.9994792342185974,0.9990740418434143],"occluded":false}],"mean_score":0.7887788414955139},{"noun_phrase":"silver rooster statue","object_id":0,"presentation_id":4,"observations":[{"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":[0.49635419249534607,0.4731481373310089,0.6890625357627869,0.7129629552364349],"occluded":false}],"mean_score":0.7887788414955139}],"dropped":{},"fates":[{"noun_phrase":"silver rooster statue","object_id":0,"frames":3,"mean_score":0.7887788414955139,"mean_area":0.03163483879810632,"outcome":"kept","presentation_id":4,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"silver rooster statue","object_id":1,"frames":3,"mean_score":0.6717948913574219,"mean_area":0.01663773167421348,"outcome":"kept","presentation_id":2,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"silver rooster statue","object_id":2,"frames":3,"mean_score":0.6800000071525574,"mean_area":0.0002584876660846839,"outcome":"kept","presentation_id":1,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"white truck","object_id":0,"frames":7,"mean_score":0.7887788414955139,"mean_area":0.03851810503982635,"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":4,"as_video":true,"fps":1.0,"max_frames":30},"outputs":{"media":"/home/zzk/tmp/rmot_smoke_real_86260/renders/--cif7Nv6BI_chunk_001_scene_010_choices.mp4","media_bytes":4296427,"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","temperature":0.0,"as_video":true,"media_bytes":4296427,"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 silver rooster ornament on top of the church roof\n2. the white truck parked on the right side of the street\n3. a person walking down the stone street","questions":["the silver rooster ornament on top of the church roof","the white truck parked on the right side of the street","a person walking down the stone street"]},"outputs":{"raw_response":"{\n  \"answers\": [\n    {\n      \"description_index\": 1,\n      \"matching_ids\": [\n        1,\n        2\n      ],\n      \"start_frame\": 1,\n      \"end_frame\": 16,\n      \"confidence\": 0.9\n    },\n    {\n      \"description_index\": 2,\n      \"matching_ids\": [\n        3\n      ],\n      \"start_frame\": 14,\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":[1,2],"start_frame":1,"end_frame":16,"confidence":0.9},{"description_index":2,"matching_ids":[3],"start_frame":14,"end_frame":16,"confidence":0.9},{"description_index":3,"matching_ids":[],"start_frame":1,"end_frame":16,"confidence":1.0}],"answers":[{"expression":"the silver rooster ornament on top of the church roof","matching_presentation_ids":[1,2],"start_choice_frame":1,"end_choice_frame":16,"confidence":0.9,"self_reported_confidence":0.9},{"expression":"the white truck parked on the right side of the street","matching_presentation_ids":[3],"start_choice_frame":14,"end_choice_frame":16,"confidence":0.9,"self_reported_confidence":0.9},{"expression":"a person walking down the stone 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":4,"selections":3,"gates":"interval, count, unknown ids, confidence, then geometry (motion)"},"outputs":{"kept":[{"expression":"the white truck parked on the right side of the street","kind":"spatial","is_negative":false,"tracks":[{"timestamps":[12.74,13.0,13.26,13.52,13.780000000000001],"boxes":[[0.8776041865348816,0.5092592239379883,0.9994791895151138,0.9722221791744232],[0.9166666865348816,0.5703703761100769,0.9994791895151138,0.9722222089767456],[0.9458333849906921,0.6314814686775208,0.9994792193174362,0.9722222089767456],[0.9447917342185974,0.6388888955116272,0.9994792379438877,0.9722222089767456],[0.9369792342185974,0.6564814448356628,0.9994792342185974,0.9990740418434143]]}]}],"rejected":{"a person walking down the stone street":"no_candidates_retrieved","the silver rooster ornament on top of the church roof":"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_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 white truck parked on the right side of the street","kind":"spatial","is_negative":false,"tracks":[{"timestamps":[12.74,13.0,13.26,13.52,13.780000000000001],"boxes":[[0.8776041865348816,0.5092592239379883,0.9994791895151138,0.9722221791744232],[0.9166666865348816,0.5703703761100769,0.9994791895151138,0.9722222089767456],[0.9458333849906921,0.6314814686775208,0.9994792193174362,0.9722222089767456],[0.9447917342185974,0.6388888955116272,0.9994792379438877,0.9722222089767456],[0.9369792342185974,0.6564814448356628,0.9994792342185974,0.9990740418434143]]}]}],"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","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","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 sign on 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 round boulder at the top of the hill\",\n      \"noun_phrase\": \"large round boulder\",\n      \"kind\": \"spatial\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the car parked on the path\",\n      \"noun_phrase\": \"car\",\n      \"kind\": \"appearance\",\n      \"expected_count\": 0,\n      \"is_negative\": true\n    }\n  ]\n}","raw_items":[{"expression":"the white sign on the stone wall","noun_phrase":"white sign","kind":"spatial","expected_count":1,"is_negative":false},{"expression":"the large round boulder at the top of the hill","noun_phrase":"large round boulder","kind":"spatial","expected_count":1,"is_negative":false},{"expression":"the car parked on the path","noun_phrase":"car","kind":"appearance","expected_count":0,"is_negative":true}],"accepted":[{"expression":"the white sign on the stone wall","noun_phrase":"white sign","kind":"spatial","expected_count":1,"is_negative":false},{"expression":"the large round boulder at the top of the hill","noun_phrase":"large round boulder","kind":"spatial","expected_count":1,"is_negative":false},{"expression":"the car parked on the path","noun_phrase":"car","kind":"appearance","expected_count":0,"is_negative":true}],"gated":[]}},"track":{"inputs":{"tracker":"Sam31MultiplexTracker","model_id":"/mnt/ceph3/ec/jacky/sam3.1","noun_phrases":["car","large round boulder","white sign"],"expressions_in":3,"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":3},"outputs":{"raw_tracks":[{"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":"white sign","object_id":0,"presentation_id":1,"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}],"mean_score":0.8552036285400391},{"noun_phrase":"white sign","object_id":1,"presentation_id":2,"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}],"mean_score":0.8632478713989258},{"noun_phrase":"white sign","object_id":3,"presentation_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}],"mean_score":0.8202247023582458},{"noun_phrase":"white sign","object_id":2,"presentation_id":4,"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}],"mean_score":0.941605806350708},{"noun_phrase":"white sign","object_id":4,"presentation_id":5,"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}],"mean_score":0.8632478713989258}],"dropped":{},"fates":[{"noun_phrase":"white sign","object_id":0,"frames":13,"mean_score":0.8552036285400391,"mean_area":0.0009850576056952638,"outcome":"kept","presentation_id":1,"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":"kept","presentation_id":2,"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":"kept","presentation_id":4,"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":"kept","presentation_id":3,"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":"kept","presentation_id":5,"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/tmp/rmot_smoke_real_86260/renders/--cif7Nv6BI_chunk_006_scene_028_choices.mp4","media_bytes":4201549,"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","temperature":0.0,"as_video":true,"media_bytes":4201549,"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 sign on the stone wall\n2. the large round boulder at the top of the hill\n3. the car parked on the path","questions":["the white sign on the stone wall","the large round boulder at the top of the hill","the car parked on the path"]},"outputs":{"raw_response":"{\n  \"answers\": [\n    {\n      \"description_index\": 1,\n      \"matching_ids\": [\n        4\n      ],\n      \"start_frame\": 1,\n      \"end_frame\": 16,\n      \"confidence\": 1.0\n    },\n    {\n      \"description_index\": 2,\n      \"matching_ids\": [],\n      \"start_frame\": 1,\n      \"end_frame\": 17,\n      \"confidence\": 0.9\n    },\n    {\n      \"description_index\": 3,\n      \"matching_ids\": [],\n      \"start_frame\": 1,\n      \"end_frame\": 17,\n      \"confidence\": 1.0\n    }\n  ]\n}","raw_answers":[{"description_index":1,"matching_ids":[4],"start_frame":1,"end_frame":16,"confidence":1.0},{"description_index":2,"matching_ids":[],"start_frame":1,"end_frame":17,"confidence":0.9},{"description_index":3,"matching_ids":[],"start_frame":1,"end_frame":17,"confidence":1.0}],"answers":[{"expression":"the white sign on the stone wall","matching_presentation_ids":[4],"start_choice_frame":1,"end_choice_frame":16,"confidence":1.0,"self_reported_confidence":1.0},{"expression":"the large round boulder at the top of the hill","matching_presentation_ids":[],"start_choice_frame":1,"end_choice_frame":17,"confidence":0.9,"self_reported_confidence":0.9},{"expression":"the car parked on the path","matching_presentation_ids":[],"start_choice_frame":1,"end_choice_frame":17,"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 sign on the stone wall","kind":"spatial","is_negative":false,"tracks":[{"timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88,10.14,10.4,10.66,10.92,11.18,11.44,11.700000000000001,11.96,12.22,12.48,12.74,13.0,13.26,13.52,13.780000000000001,14.040000000000001,14.3,14.56,14.82,15.08],"boxes":[[0.3343750238418579,0.5537036657333374,0.38854169473052025,0.6564814448356628],[0.34010419249534607,0.5888888835906982,0.39479169622063637,0.6916666626930237],[0.3348958492279053,0.5694444179534912,0.3916666842997074,0.6851851567626],[0.3395833373069763,0.5379629731178284,0.39635417237877846,0.6574074178934097],[0.33125001192092896,0.5490740537643433,0.3916666805744171,0.6712962761521339],[0.28697916865348816,0.5361111164093018,0.3526041731238365,0.6537037044763565],[0.26510417461395264,0.5120370388031006,0.3322916775941849,0.6435185223817825],[0.29375001788139343,0.614814817905426,0.36093752086162567,0.7407407462596893],[0.3411458432674408,0.5944444537162781,0.4062500149011612,0.7268518656492233],[0.3541666865348816,0.6018518209457397,0.4203125238418579,0.7407407015562057],[0.3864583671092987,0.5805555582046509,0.4500000402331352,0.7018518522381783],[0.39531251788139343,0.5777777433395386,0.46041668951511383,0.6999999657273293],[0.4270833432674408,0.614814817905426,0.4932291805744171,0.7453703731298447],[0.4244791865348816,0.6240740418434143,0.49375002086162567,0.7564814537763596],[0.3786458671092987,0.7342592477798462,0.45364587008953094,0.8935185074806213],[0.3463541865348816,0.6129629611968994,0.48854169249534607,0.9148148000240326],[0.3656250238418579,0.7694444060325623,0.4463541954755783,0.9268518090248108],[0.004687500186264515,0.6370370388031006,0.4515625359490514,0.9185185134410858],[0.004687500186264515,0.5537036657333374,0.4395833434537053,0.9490740299224854],[0.3140625059604645,0.7509258985519409,0.4072916805744171,0.9249999672174454],[0.3177083432674408,0.7351851463317871,0.41197918355464935,0.9120370000600815],[0.3177083432674408,0.7648147940635681,0.4151041805744171,0.9490740448236465],[0.3177083432674408,0.7407407164573669,0.41718751192092896,0.9240740537643433],[0.30156251788139343,0.7138888835906982,0.40781252086162567,0.9074074029922485],[0.26093751192092896,0.6685184836387634,0.37291668355464935,0.866666629910469],[0.29635417461395264,0.6833333373069763,0.4052083492279053,0.8777777850627899],[0.3239583373069763,0.7240740656852722,0.42916667461395264,0.9166666567325592],[0.3229166865348816,0.7379629611968994,0.43177086114883423,0.9351851791143417],[0.17135417461395264,0.8138888478279114,0.4125000238418579,0.9990740269422531],[0.18541668355464935,0.6712962985038757,0.45625002682209015,0.9990740716457367],[0.24114584922790527,0.8083333373069763,0.45364585518836975,0.9990740716457367],[0.2828125059604645,0.8388888835906982,0.518750011920929,0.9990740716457367],[0.4255208671092987,0.8064814805984497,0.5322917029261589,0.9722222238779068],[0.4130208492279053,0.7990740537643433,0.5744791924953461,0.9990740567445755],[0.06041666865348816,0.8166666626930237,0.5088541805744171,0.9990740716457367],[0.32447919249534607,0.7944444417953491,0.476041704416275,0.9990740716457367],[0.30000001192092896,0.815740704536438,0.450520858168602,0.9990740418434143],[0.24843750894069672,0.8583332896232605,0.41979168355464935,0.9990740269422531],[0.20156250894069672,0.8888888955116272,0.39270834624767303,0.9722222238779068],[0.14166668057441711,0.8518518209457397,0.35104168951511383,0.9722221866250038],[0.09479167312383652,0.8787037134170532,0.31458335369825363,0.9722222313284874],[0.08958333730697632,0.8796296119689941,0.3177083432674408,0.972222201526165],[0.08645834028720856,0.8638888597488403,0.31718751788139343,0.9722221940755844],[0.09218750149011612,0.8388888835906982,0.3244791850447655,0.9722222089767456],[0.08229167014360428,0.8361110687255859,0.3161458447575569,0.9722221791744232],[0.09687500447034836,0.8694444298744202,0.32500001043081284,0.9722222089767456],[0.07864584028720856,0.8787037134170532,0.30885419249534607,0.9722222313284874],[0.06927083432674408,0.8879629373550415,0.2994791865348816,0.9722221940755844],[0.06875000149011612,0.895370364189148,0.2984375134110451,0.9722222164273262],[0.07864584028720856,0.8972222208976746,0.3072916865348816,0.9731481447815895],[0.07708333432674408,0.8944444060325623,0.30520834028720856,0.9722221791744232],[0.06822916865348816,0.9111111164093018,0.2776041775941849,0.9722222276031971],[0.06979166716337204,0.9435185194015503,0.17239584028720856,0.9722222220152617],[0.07604166865348816,0.9342592358589172,0.1979166716337204,0.9722221978008747],[0.07604166865348816,0.9259259104728699,0.2213541716337204,0.9722222052514553],[0.07916667312383652,0.9351851940155029,0.19687501341104507,0.9722222313284874],[0.08437500149011612,0.9351851940155029,0.213541679084301,0.9722222313284874],[0.10208334028720856,0.9268518090248108,0.28333334624767303,0.9722221791744232],[0.12343750894069672,0.9101851582527161,0.33020834624767303,0.9722221940755844]]}]}],"rejected":{"the car parked on the path":"no_candidates_retrieved","the large round boulder at the top of the hill":"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_006_scene_028.mp4","scene_key":"--cif7Nv6BI_chunk_006_scene_028","category":"test_vehicles","height":1080,"width":1920,"fps":50.0,"num_frames":807,"duration":16.14,"rmot_expressions":[{"expression":"the white sign on the stone wall","kind":"spatial","is_negative":false,"tracks":[{"timestamps":[0.0,0.26,0.52,0.78,1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88,10.14,10.4,10.66,10.92,11.18,11.44,11.700000000000001,11.96,12.22,12.48,12.74,13.0,13.26,13.52,13.780000000000001,14.040000000000001,14.3,14.56,14.82,15.08],"boxes":[[0.3343750238418579,0.5537036657333374,0.38854169473052025,0.6564814448356628],[0.34010419249534607,0.5888888835906982,0.39479169622063637,0.6916666626930237],[0.3348958492279053,0.5694444179534912,0.3916666842997074,0.6851851567626],[0.3395833373069763,0.5379629731178284,0.39635417237877846,0.6574074178934097],[0.33125001192092896,0.5490740537643433,0.3916666805744171,0.6712962761521339],[0.28697916865348816,0.5361111164093018,0.3526041731238365,0.6537037044763565],[0.26510417461395264,0.5120370388031006,0.3322916775941849,0.6435185223817825],[0.29375001788139343,0.614814817905426,0.36093752086162567,0.7407407462596893],[0.3411458432674408,0.5944444537162781,0.4062500149011612,0.7268518656492233],[0.3541666865348816,0.6018518209457397,0.4203125238418579,0.7407407015562057],[0.3864583671092987,0.5805555582046509,0.4500000402331352,0.7018518522381783],[0.39531251788139343,0.5777777433395386,0.46041668951511383,0.6999999657273293],[0.4270833432674408,0.614814817905426,0.4932291805744171,0.7453703731298447],[0.4244791865348816,0.6240740418434143,0.49375002086162567,0.7564814537763596],[0.3786458671092987,0.7342592477798462,0.45364587008953094,0.8935185074806213],[0.3463541865348816,0.6129629611968994,0.48854169249534607,0.9148148000240326],[0.3656250238418579,0.7694444060325623,0.4463541954755783,0.9268518090248108],[0.004687500186264515,0.6370370388031006,0.4515625359490514,0.9185185134410858],[0.004687500186264515,0.5537036657333374,0.4395833434537053,0.9490740299224854],[0.3140625059604645,0.7509258985519409,0.4072916805744171,0.9249999672174454],[0.3177083432674408,0.7351851463317871,0.41197918355464935,0.9120370000600815],[0.3177083432674408,0.7648147940635681,0.4151041805744171,0.9490740448236465],[0.3177083432674408,0.7407407164573669,0.41718751192092896,0.9240740537643433],[0.30156251788139343,0.7138888835906982,0.40781252086162567,0.9074074029922485],[0.26093751192092896,0.6685184836387634,0.37291668355464935,0.866666629910469],[0.29635417461395264,0.6833333373069763,0.4052083492279053,0.8777777850627899],[0.3239583373069763,0.7240740656852722,0.42916667461395264,0.9166666567325592],[0.3229166865348816,0.7379629611968994,0.43177086114883423,0.9351851791143417],[0.17135417461395264,0.8138888478279114,0.4125000238418579,0.9990740269422531],[0.18541668355464935,0.6712962985038757,0.45625002682209015,0.9990740716457367],[0.24114584922790527,0.8083333373069763,0.45364585518836975,0.9990740716457367],[0.2828125059604645,0.8388888835906982,0.518750011920929,0.9990740716457367],[0.4255208671092987,0.8064814805984497,0.5322917029261589,0.9722222238779068],[0.4130208492279053,0.7990740537643433,0.5744791924953461,0.9990740567445755],[0.06041666865348816,0.8166666626930237,0.5088541805744171,0.9990740716457367],[0.32447919249534607,0.7944444417953491,0.476041704416275,0.9990740716457367],[0.30000001192092896,0.815740704536438,0.450520858168602,0.9990740418434143],[0.24843750894069672,0.8583332896232605,0.41979168355464935,0.9990740269422531],[0.20156250894069672,0.8888888955116272,0.39270834624767303,0.9722222238779068],[0.14166668057441711,0.8518518209457397,0.35104168951511383,0.9722221866250038],[0.09479167312383652,0.8787037134170532,0.31458335369825363,0.9722222313284874],[0.08958333730697632,0.8796296119689941,0.3177083432674408,0.972222201526165],[0.08645834028720856,0.8638888597488403,0.31718751788139343,0.9722221940755844],[0.09218750149011612,0.8388888835906982,0.3244791850447655,0.9722222089767456],[0.08229167014360428,0.8361110687255859,0.3161458447575569,0.9722221791744232],[0.09687500447034836,0.8694444298744202,0.32500001043081284,0.9722222089767456],[0.07864584028720856,0.8787037134170532,0.30885419249534607,0.9722222313284874],[0.06927083432674408,0.8879629373550415,0.2994791865348816,0.9722221940755844],[0.06875000149011612,0.895370364189148,0.2984375134110451,0.9722222164273262],[0.07864584028720856,0.8972222208976746,0.3072916865348816,0.9731481447815895],[0.07708333432674408,0.8944444060325623,0.30520834028720856,0.9722221791744232],[0.06822916865348816,0.9111111164093018,0.2776041775941849,0.9722222276031971],[0.06979166716337204,0.9435185194015503,0.17239584028720856,0.9722222220152617],[0.07604166865348816,0.9342592358589172,0.1979166716337204,0.9722221978008747],[0.07604166865348816,0.9259259104728699,0.2213541716337204,0.9722222052514553],[0.07916667312383652,0.9351851940155029,0.19687501341104507,0.9722222313284874],[0.08437500149011612,0.9351851940155029,0.213541679084301,0.9722222313284874],[0.10208334028720856,0.9268518090248108,0.28333334624767303,0.9722221791744232],[0.12343750894069672,0.9101851582527161,0.33020834624767303,0.9722221940755844]]}]}],"rmot_expression_count":1,"rmot_negative_count":0,"rmot_motion_count":0}}}}},{"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","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 with long dark hair standing in front of the shop\",\n      \"noun_phrase\": \"woman with long dark hair\",\n      \"kind\": \"appearance\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the hand that waves on the left side of the screen\",\n      \"noun_phrase\": \"hand\",\n      \"kind\": \"motion\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the man standing next to the woman\",\n      \"noun_phrase\": \"man\",\n      \"kind\": \"spatial\",\n      \"expected_count\": 0,\n      \"is_negative\": true\n    }\n  ]\n}","raw_items":[{"expression":"the woman with long dark hair standing in front of the shop","noun_phrase":"woman with long dark hair","kind":"appearance","expected_count":1,"is_negative":false},{"expression":"the hand that waves on the left side of the screen","noun_phrase":"hand","kind":"motion","expected_count":1,"is_negative":false},{"expression":"the man standing next to the woman","noun_phrase":"man","kind":"spatial","expected_count":0,"is_negative":true}],"accepted":[{"expression":"the woman with long dark hair standing in front of the shop","noun_phrase":"woman with long dark hair","kind":"appearance","expected_count":1,"is_negative":false},{"expression":"the hand that waves on the left side of the screen","noun_phrase":"hand","kind":"motion","expected_count":1,"is_negative":false},{"expression":"the man standing next to the woman","noun_phrase":"man","kind":"spatial","expected_count":0,"is_negative":true}],"gated":[]}},"track":{"inputs":{"tracker":"Sam31MultiplexTracker","model_id":"/mnt/ceph3/ec/jacky/sam3.1","noun_phrases":["hand","man","woman with long dark hair"],"expressions_in":3,"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":3},"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 with long dark hair","object_id":0,"observations":[{"timestamp":0.0,"box":[0.3005208373069763,0.17129629850387573,0.7781250178813934,0.970370352268219],"occluded":false},{"timestamp":0.26,"box":[0.30937501788139343,0.1787036955356598,0.7463541924953461,0.9722221791744232],"occluded":false},{"timestamp":0.52,"box":[0.29375001788139343,0.182407408952713,0.7125000357627869,0.972222238779068],"occluded":false},{"timestamp":0.78,"box":[0.3135416805744171,0.19814814627170563,0.7161458730697632,0.9740740805864334],"occluded":false},{"timestamp":1.04,"box":[0.11562500894069672,0.23148147761821747,0.7416667193174362,0.9731481522321701],"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":[0.3708333671092987,0.15925925970077515,0.7895833849906921,0.9712963104248047],"occluded":false},{"timestamp":3.9,"box":[0.3619791865348816,0.15648147463798523,0.7791666984558105,0.9731481373310089],"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":[0.28177085518836975,0.1814814805984497,0.6765625476837158,0.9740740656852722],"occluded":false},{"timestamp":5.46,"box":[0.2760416865348816,0.18703703582286835,0.6755208671092987,0.9750000089406967],"occluded":false},{"timestamp":5.72,"box":[0.2578125,0.19074073433876038,0.6744791865348816,0.9731481373310089],"occluded":false},{"timestamp":5.98,"box":[0.26979169249534607,0.19629628956317902,0.6817708909511566,0.9749999791383743],"occluded":false},{"timestamp":6.24,"box":[0.2828125059604645,0.192592591047287,0.694791704416275,0.9749999940395355],"occluded":false},{"timestamp":6.5,"box":[0.2968750298023224,0.17685185372829437,0.7156250476837158,0.9722221940755844],"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":[0.29375001788139343,0.14351852238178253,0.7364583611488342,0.9722222238779068],"occluded":false},{"timestamp":7.54,"box":[0.30156251788139343,0.13518518209457397,0.7473958730697632,0.9722222089767456],"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":[0.30781251192092896,0.1509259194135666,0.7468750476837158,0.9722221940755844],"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":[0.29114586114883423,0.14351852238178253,0.7312500476837158,0.9722222238779068],"occluded":false},{"timestamp":9.36,"box":[0.2786458432674408,0.14537036418914795,0.7151041924953461,0.9722222089767456],"occluded":false},{"timestamp":9.620000000000001,"box":[0.2734375,0.15555554628372192,0.7026041746139526,0.9722222089767456],"occluded":false},{"timestamp":9.88,"box":[0.2614583373069763,0.1537037044763565,0.6895833611488342,0.9731481224298477],"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":[0.24687501788139343,0.16111110150814056,0.6760416924953461,0.9768518060445786],"occluded":false},{"timestamp":10.92,"box":[0.2473958432674408,0.18333333730697632,0.6802083551883698,0.9731481671333313],"occluded":false},{"timestamp":11.18,"box":[0.24635417759418488,0.16296295821666718,0.6833333522081375,0.9731481522321701],"occluded":false},{"timestamp":11.44,"box":[0.2343750149011612,0.14814814925193787,0.680208370089531,0.9740740358829498],"occluded":false},{"timestamp":11.700000000000001,"box":[0.26875001192092896,0.16388888657093048,0.7161458730697632,0.9731481224298477],"occluded":false},{"timestamp":11.96,"box":[0.2671875059604645,0.1787036955356598,0.7140625417232513,0.9731481373310089],"occluded":false},{"timestamp":12.22,"box":[0.2723958492279053,0.17685185372829437,0.7161458730697632,0.9731481522321701],"occluded":false},{"timestamp":12.48,"box":[0.2848958373069763,0.182407408952713,0.7265625298023224,0.9731481373310089],"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":[0.4208333492279053,0.1379629671573639,0.8656250536441803,0.972222238779068],"occluded":false},{"timestamp":13.780000000000001,"box":[0.3552083373069763,0.17592592537403107,0.7588541805744171,0.9722222238779068],"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":[0.09218750149011612,0.10462962836027145,0.5416666939854622,0.972222201526165],"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":[0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116,0.8904109597206116]}],"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 with long dark hair","object_id":0,"presentation_id":2,"observations":[{"timestamp":0.0,"box":[0.3005208373069763,0.17129629850387573,0.7781250178813934,0.970370352268219],"occluded":false},{"timestamp":0.26,"box":[0.30937501788139343,0.1787036955356598,0.7463541924953461,0.9722221791744232],"occluded":false},{"timestamp":0.52,"box":[0.29375001788139343,0.182407408952713,0.7125000357627869,0.972222238779068],"occluded":false},{"timestamp":0.78,"box":[0.3135416805744171,0.19814814627170563,0.7161458730697632,0.9740740805864334],"occluded":false},{"timestamp":1.04,"box":[0.11562500894069672,0.23148147761821747,0.7416667193174362,0.9731481522321701],"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":[0.3708333671092987,0.15925925970077515,0.7895833849906921,0.9712963104248047],"occluded":false},{"timestamp":3.9,"box":[0.3619791865348816,0.15648147463798523,0.7791666984558105,0.9731481373310089],"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":[0.28177085518836975,0.1814814805984497,0.6765625476837158,0.9740740656852722],"occluded":false},{"timestamp":5.46,"box":[0.2760416865348816,0.18703703582286835,0.6755208671092987,0.9750000089406967],"occluded":false},{"timestamp":5.72,"box":[0.2578125,0.19074073433876038,0.6744791865348816,0.9731481373310089],"occluded":false},{"timestamp":5.98,"box":[0.26979169249534607,0.19629628956317902,0.6817708909511566,0.9749999791383743],"occluded":false},{"timestamp":6.24,"box":[0.2828125059604645,0.192592591047287,0.694791704416275,0.9749999940395355],"occluded":false},{"timestamp":6.5,"box":[0.2968750298023224,0.17685185372829437,0.7156250476837158,0.9722221940755844],"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":[0.29375001788139343,0.14351852238178253,0.7364583611488342,0.9722222238779068],"occluded":false},{"timestamp":7.54,"box":[0.30156251788139343,0.13518518209457397,0.7473958730697632,0.9722222089767456],"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":[0.30781251192092896,0.1509259194135666,0.7468750476837158,0.9722221940755844],"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":[0.29114586114883423,0.14351852238178253,0.7312500476837158,0.9722222238779068],"occluded":false},{"timestamp":9.36,"box":[0.2786458432674408,0.14537036418914795,0.7151041924953461,0.9722222089767456],"occluded":false},{"timestamp":9.620000000000001,"box":[0.2734375,0.15555554628372192,0.7026041746139526,0.9722222089767456],"occluded":false},{"timestamp":9.88,"box":[0.2614583373069763,0.1537037044763565,0.6895833611488342,0.9731481224298477],"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":[0.24687501788139343,0.16111110150814056,0.6760416924953461,0.9768518060445786],"occluded":false},{"timestamp":10.92,"box":[0.2473958432674408,0.18333333730697632,0.6802083551883698,0.9731481671333313],"occluded":false},{"timestamp":11.18,"box":[0.24635417759418488,0.16296295821666718,0.6833333522081375,0.9731481522321701],"occluded":false},{"timestamp":11.44,"box":[0.2343750149011612,0.14814814925193787,0.680208370089531,0.9740740358829498],"occluded":false},{"timestamp":11.700000000000001,"box":[0.26875001192092896,0.16388888657093048,0.7161458730697632,0.9731481224298477],"occluded":false},{"timestamp":11.96,"box":[0.2671875059604645,0.1787036955356598,0.7140625417232513,0.9731481373310089],"occluded":false},{"timestamp":12.22,"box":[0.2723958492279053,0.17685185372829437,0.7161458730697632,0.9731481522321701],"occluded":false},{"timestamp":12.48,"box":[0.2848958373069763,0.182407408952713,0.7265625298023224,0.9731481373310089],"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":[0.4208333492279053,0.1379629671573639,0.8656250536441803,0.972222238779068],"occluded":false},{"timestamp":13.780000000000001,"box":[0.3552083373069763,0.17592592537403107,0.7588541805744171,0.9722222238779068],"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":[0.09218750149011612,0.10462962836027145,0.5416666939854622,0.972222201526165],"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":0.8904109597206116}],"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 with long dark hair","object_id":0,"frames":59,"mean_score":0.8904109597206116,"mean_area":0.34486007757173087,"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/tmp/rmot_smoke_real_86260/renders/--cif7Nv6BI_chunk_014_scene_013_choices.mp4","media_bytes":5677179,"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","temperature":0.0,"as_video":true,"media_bytes":5677179,"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 with long dark hair standing in front of the shop\n2. the hand that waves on the left side of the screen\n3. the man standing next to the woman","questions":["the woman with long dark hair standing in front of the shop","the hand that waves on the left side of the screen","the man standing next to the woman"]},"outputs":{"raw_response":"{\n  \"answers\": [\n    {\n      \"description_index\": 1,\n      \"matching_ids\": [2],\n      \"start_frame\": 1,\n      \"end_frame\": 16,\n      \"confidence\": 1.0\n    },\n    {\n      \"description_index\": 2,\n      \"matching_ids\": [1],\n      \"start_frame\": 2,\n      \"end_frame\": 3,\n      \"confidence\": 1.0\n    },\n    {\n      \"description_index\": 3,\n      \"matching_ids\": [],\n      \"start_frame\": 1,\n      \"end_frame\": 19,\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},{"description_index":3,"matching_ids":[],"start_frame":1,"end_frame":19,"confidence":1.0}],"answers":[{"expression":"the woman with long dark hair standing 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 that waves on the left side of the screen","matching_presentation_ids":[1],"start_choice_frame":2,"end_choice_frame":3,"confidence":1.0,"self_reported_confidence":1.0},{"expression":"the man standing next to the woman","matching_presentation_ids":[],"start_choice_frame":1,"end_choice_frame":19,"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 woman with long dark hair standing 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":[[0.3005208373069763,0.17129629850387573,0.7781250178813934,0.970370352268219],[0.30937501788139343,0.1787036955356598,0.7463541924953461,0.9722221791744232],[0.29375001788139343,0.182407408952713,0.7125000357627869,0.972222238779068],[0.3135416805744171,0.19814814627170563,0.7161458730697632,0.9740740805864334],[0.11562500894069672,0.23148147761821747,0.7416667193174362,0.9731481522321701],[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],[0.3708333671092987,0.15925925970077515,0.7895833849906921,0.9712963104248047],[0.3619791865348816,0.15648147463798523,0.7791666984558105,0.9731481373310089],[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],[0.28177085518836975,0.1814814805984497,0.6765625476837158,0.9740740656852722],[0.2760416865348816,0.18703703582286835,0.6755208671092987,0.9750000089406967],[0.2578125,0.19074073433876038,0.6744791865348816,0.9731481373310089],[0.26979169249534607,0.19629628956317902,0.6817708909511566,0.9749999791383743],[0.2828125059604645,0.192592591047287,0.694791704416275,0.9749999940395355],[0.2968750298023224,0.17685185372829437,0.7156250476837158,0.9722221940755844],[0.30156251788139343,0.1657407432794571,0.7281250357627869,0.9731481224298477],[0.29479166865348816,0.14999999105930328,0.7270833551883698,0.9722222238779068],[0.29375001788139343,0.14351852238178253,0.7364583611488342,0.9722222238779068],[0.30156251788139343,0.13518518209457397,0.7473958730697632,0.9722222089767456],[0.3083333373069763,0.14999999105930328,0.7562500238418579,0.9731481224298477],[0.30781251192092896,0.1509259194135666,0.7520833611488342,0.9722221940755844],[0.30781251192092896,0.1509259194135666,0.7468750476837158,0.9722221940755844],[0.29635417461395264,0.13611111044883728,0.7437500357627869,0.9722221791744232],[0.2942708432674408,0.138888880610466,0.7385416924953461,0.9722221940755844],[0.29114586114883423,0.14351852238178253,0.7312500476837158,0.9722222238779068],[0.2786458432674408,0.14537036418914795,0.7151041924953461,0.9722222089767456],[0.2734375,0.15555554628372192,0.7026041746139526,0.9722222089767456],[0.2614583373069763,0.1537037044763565,0.6895833611488342,0.9731481224298477],[0.2526041865348816,0.1537037044763565,0.6786458790302277,0.9731481224298477],[0.25052085518836975,0.14537036418914795,0.6812500357627869,0.9722222089767456],[0.24687501788139343,0.16111110150814056,0.6760416924953461,0.9768518060445786],[0.2473958432674408,0.18333333730697632,0.6802083551883698,0.9731481671333313],[0.24635417759418488,0.16296295821666718,0.6833333522081375,0.9731481522321701],[0.2343750149011612,0.14814814925193787,0.680208370089531,0.9740740358829498],[0.26875001192092896,0.16388888657093048,0.7161458730697632,0.9731481224298477],[0.2671875059604645,0.1787036955356598,0.7140625417232513,0.9731481373310089],[0.2723958492279053,0.17685185372829437,0.7161458730697632,0.9731481522321701],[0.2848958373069763,0.182407408952713,0.7265625298023224,0.9731481373310089],[0.3187500238418579,0.18981480598449707,0.7656250596046448,0.9740740656852722],[0.3609375059604645,0.1527777761220932,0.8343750238418579,0.9712962955236435],[0.41979169845581055,0.13333332538604736,0.8947917222976685,0.9722222089767456],[0.4208333492279053,0.1379629671573639,0.8656250536441803,0.972222238779068],[0.3552083373069763,0.17592592537403107,0.7588541805744171,0.9722222238779068],[0.2526041865348816,0.182407408952713,0.6328125298023224,0.972222238779068],[0.1302083432674408,0.1527777761220932,0.5536458790302277,0.9722221940755844],[0.09218750149011612,0.10462962836027145,0.5416666939854622,0.972222201526165],[0.0,0.028703702613711357,0.4427083432674408,0.971296263858676]]}]},{"expression":"the hand that waves on the left side of the screen","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":{"the man standing next to the woman":"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_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 with long dark hair standing 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":[[0.3005208373069763,0.17129629850387573,0.7781250178813934,0.970370352268219],[0.30937501788139343,0.1787036955356598,0.7463541924953461,0.9722221791744232],[0.29375001788139343,0.182407408952713,0.7125000357627869,0.972222238779068],[0.3135416805744171,0.19814814627170563,0.7161458730697632,0.9740740805864334],[0.11562500894069672,0.23148147761821747,0.7416667193174362,0.9731481522321701],[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],[0.3708333671092987,0.15925925970077515,0.7895833849906921,0.9712963104248047],[0.3619791865348816,0.15648147463798523,0.7791666984558105,0.9731481373310089],[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],[0.28177085518836975,0.1814814805984497,0.6765625476837158,0.9740740656852722],[0.2760416865348816,0.18703703582286835,0.6755208671092987,0.9750000089406967],[0.2578125,0.19074073433876038,0.6744791865348816,0.9731481373310089],[0.26979169249534607,0.19629628956317902,0.6817708909511566,0.9749999791383743],[0.2828125059604645,0.192592591047287,0.694791704416275,0.9749999940395355],[0.2968750298023224,0.17685185372829437,0.7156250476837158,0.9722221940755844],[0.30156251788139343,0.1657407432794571,0.7281250357627869,0.9731481224298477],[0.29479166865348816,0.14999999105930328,0.7270833551883698,0.9722222238779068],[0.29375001788139343,0.14351852238178253,0.7364583611488342,0.9722222238779068],[0.30156251788139343,0.13518518209457397,0.7473958730697632,0.9722222089767456],[0.3083333373069763,0.14999999105930328,0.7562500238418579,0.9731481224298477],[0.30781251192092896,0.1509259194135666,0.7520833611488342,0.9722221940755844],[0.30781251192092896,0.1509259194135666,0.7468750476837158,0.9722221940755844],[0.29635417461395264,0.13611111044883728,0.7437500357627869,0.9722221791744232],[0.2942708432674408,0.138888880610466,0.7385416924953461,0.9722221940755844],[0.29114586114883423,0.14351852238178253,0.7312500476837158,0.9722222238779068],[0.2786458432674408,0.14537036418914795,0.7151041924953461,0.9722222089767456],[0.2734375,0.15555554628372192,0.7026041746139526,0.9722222089767456],[0.2614583373069763,0.1537037044763565,0.6895833611488342,0.9731481224298477],[0.2526041865348816,0.1537037044763565,0.6786458790302277,0.9731481224298477],[0.25052085518836975,0.14537036418914795,0.6812500357627869,0.9722222089767456],[0.24687501788139343,0.16111110150814056,0.6760416924953461,0.9768518060445786],[0.2473958432674408,0.18333333730697632,0.6802083551883698,0.9731481671333313],[0.24635417759418488,0.16296295821666718,0.6833333522081375,0.9731481522321701],[0.2343750149011612,0.14814814925193787,0.680208370089531,0.9740740358829498],[0.26875001192092896,0.16388888657093048,0.7161458730697632,0.9731481224298477],[0.2671875059604645,0.1787036955356598,0.7140625417232513,0.9731481373310089],[0.2723958492279053,0.17685185372829437,0.7161458730697632,0.9731481522321701],[0.2848958373069763,0.182407408952713,0.7265625298023224,0.9731481373310089],[0.3187500238418579,0.18981480598449707,0.7656250596046448,0.9740740656852722],[0.3609375059604645,0.1527777761220932,0.8343750238418579,0.9712962955236435],[0.41979169845581055,0.13333332538604736,0.8947917222976685,0.9722222089767456],[0.4208333492279053,0.1379629671573639,0.8656250536441803,0.972222238779068],[0.3552083373069763,0.17592592537403107,0.7588541805744171,0.9722222238779068],[0.2526041865348816,0.182407408952713,0.6328125298023224,0.972222238779068],[0.1302083432674408,0.1527777761220932,0.5536458790302277,0.9722221940755844],[0.09218750149011612,0.10462962836027145,0.5416666939854622,0.972222201526165],[0.0,0.028703702613711357,0.4427083432674408,0.971296263858676]]}]},{"expression":"the hand that waves on the left side of the screen","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","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 talking and gesturing with his hands\",\n      \"noun_phrase\": \"man\",\n      \"kind\": \"motion\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the green and black cap worn by the man\",\n      \"noun_phrase\": \"cap\",\n      \"kind\": \"appearance\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the blue baseball cap\",\n      \"noun_phrase\": \"cap\",\n      \"kind\": \"appearance\",\n      \"expected_count\": 0,\n      \"is_negative\": true\n    }\n  ]\n}","raw_items":[{"expression":"the man talking and gesturing with his hands","noun_phrase":"man","kind":"motion","expected_count":1,"is_negative":false},{"expression":"the green and black cap worn by the man","noun_phrase":"cap","kind":"appearance","expected_count":1,"is_negative":false},{"expression":"the blue baseball cap","noun_phrase":"cap","kind":"appearance","expected_count":0,"is_negative":true}],"accepted":[{"expression":"the man talking and gesturing with his hands","noun_phrase":"man","kind":"motion","expected_count":1,"is_negative":false},{"expression":"the green and black cap worn by the man","noun_phrase":"cap","kind":"appearance","expected_count":1,"is_negative":false},{"expression":"the blue baseball cap","noun_phrase":"cap","kind":"appearance","expected_count":0,"is_negative":true}],"gated":[]}},"track":{"inputs":{"tracker":"Sam31MultiplexTracker","model_id":"/mnt/ceph3/ec/jacky/sam3.1","noun_phrases":["cap","man"],"expressions_in":3,"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":2},"outputs":{"raw_tracks":[{"noun_phrase":"cap","object_id":0,"observations":[{"timestamp":0.0,"box":[0.4062500298023224,0.014814814552664757,0.679166704416275,0.34722222946584225],"occluded":false},{"timestamp":0.25025,"box":[0.4270833432674408,0.014814814552664757,0.7151041924953461,0.5120370294898748],"occluded":false},{"timestamp":0.5005,"box":[0.4411458671092987,0.02222222089767456,0.736979216337204,0.3574073910713196],"occluded":false},{"timestamp":0.75075,"box":[0.4359375238418579,0.023148147389292717,0.7286458611488342,0.5537036936730146],"occluded":false},{"timestamp":1.001,"box":[0.4203125238418579,0.023148147389292717,0.7020833790302277,0.5259259026497602],"occluded":false},{"timestamp":1.25125,"box":[0.40416669845581055,0.021296296268701553,0.6703125536441803,0.5259259082376957],"occluded":false},{"timestamp":1.5015,"box":[0.39479169249534607,0.013888888992369175,0.6630208790302277,0.595370345748961],"occluded":false},{"timestamp":1.75175,"box":[0.3838541805744171,0.018518518656492233,0.6536458730697632,0.5537036769092083],"occluded":false},{"timestamp":2.002,"box":[0.3864583671092987,0.023148147389292717,0.6567708849906921,0.5675925891846418],"occluded":false},{"timestamp":2.25225,"box":[0.3958333432674408,0.01944444328546524,0.6697916984558105,0.5398148074746132],"occluded":false},{"timestamp":2.5025,"box":[0.3958333432674408,0.018518518656492233,0.6729166805744171,0.6092592589557171],"occluded":false},{"timestamp":2.75275,"box":[0.39635419845581055,0.020370369777083397,0.674479216337204,0.5537036713212729],"occluded":false},{"timestamp":3.003,"box":[0.40052086114883423,0.018518518656492233,0.6807292103767395,0.5398147813975811],"occluded":false},{"timestamp":3.25325,"box":[0.4062500298023224,0.010185184888541698,0.6848958730697632,0.5537036685273051],"occluded":false},{"timestamp":3.5035,"box":[0.40781253576278687,0.01111111044883728,0.6848958730697632,0.4564814865589142],"occluded":false},{"timestamp":3.7537499999999997,"box":[0.4088541865348816,0.015740741044282913,0.686979204416275,0.4981481321156025],"occluded":false},{"timestamp":4.004,"box":[0.4052083492279053,0.026851851493120193,0.679166704416275,0.36574073508381844],"occluded":false},{"timestamp":4.25425,"box":[0.4036458432674408,0.024074073880910873,0.6802083551883698,0.4842592440545559],"occluded":false},{"timestamp":4.5045,"box":[0.40156251192092896,0.02222222089767456,0.6802083551883698,0.470370352268219],"occluded":false},{"timestamp":4.75475,"box":[0.40000003576278687,0.024074073880910873,0.6781250536441803,0.4842592440545559],"occluded":false},{"timestamp":5.005,"box":[0.3994791805744171,0.02500000037252903,0.6744791865348816,0.5398147944360971],"occluded":false},{"timestamp":5.25525,"box":[0.3984375298023224,0.024074073880910873,0.6734375357627869,0.5398148261010647],"occluded":false},{"timestamp":5.5055,"box":[0.3973958492279053,0.024074073880910873,0.6718750298023224,0.3583333157002926],"occluded":false},{"timestamp":5.75575,"box":[0.3984375298023224,0.024074073880910873,0.6708333790302277,0.5537037216126919],"occluded":false},{"timestamp":6.006,"box":[0.37552085518836975,0.017592592164874077,0.6567708551883698,0.5814814995974302],"occluded":false},{"timestamp":6.25625,"box":[0.33020836114883423,0.023148147389292717,0.6447917222976685,0.5675925891846418],"occluded":false},{"timestamp":6.5065,"box":[0.32604169845581055,0.025925925001502037,0.6411458849906921,0.581481447443366],"occluded":false},{"timestamp":6.756749999999999,"box":[0.3395833373069763,0.024074073880910873,0.6442708671092987,0.4148147962987423],"occluded":false},{"timestamp":7.007,"box":[0.3593750298023224,0.02222222089767456,0.6427083909511566,0.539814829826355],"occluded":false},{"timestamp":7.25725,"box":[0.36770835518836975,0.02500000037252903,0.6536458730697632,0.5120370332151651],"occluded":false},{"timestamp":7.507499999999999,"box":[0.36770835518836975,0.018518518656492233,0.6661458611488342,0.5814814679324627],"occluded":false},{"timestamp":7.75775,"box":[0.3760416805744171,0.017592592164874077,0.6677083671092987,0.45648146979510784],"occluded":false},{"timestamp":8.008,"box":[0.3765625059604645,0.012037036940455437,0.6645833551883698,0.369444428011775],"occluded":false},{"timestamp":8.25825,"box":[0.3760416805744171,0.0,0.6598958671092987,0.4842592477798462],"occluded":false},{"timestamp":8.5085,"box":[0.3786458671092987,0.0,0.6619792282581329,0.4981481432914734],"occluded":false},{"timestamp":8.75875,"box":[0.3765625059604645,0.0,0.6552083492279053,0.4009259045124054],"occluded":false},{"timestamp":9.009,"box":[0.37552085518836975,0.009259259328246117,0.6500000357627869,0.5537037011235952],"occluded":false},{"timestamp":9.25925,"box":[0.37031251192092896,0.018518518656492233,0.6453125178813934,0.372222226113081],"occluded":false},{"timestamp":9.5095,"box":[0.37031251192092896,0.023148147389292717,0.6442708671092987,0.37407406978309155],"occluded":false},{"timestamp":9.75975,"box":[0.3697916865348816,0.026851851493120193,0.647916704416275,0.5675925798714161],"occluded":false},{"timestamp":10.01,"box":[0.37031251192092896,0.023148147389292717,0.6500000357627869,0.49814814142882824],"occluded":false},{"timestamp":10.26025,"box":[0.37812501192092896,0.023148147389292717,0.6567708551883698,0.5120370369404554],"occluded":false},{"timestamp":10.5105,"box":[0.37916669249534607,0.02500000037252903,0.6546875536441803,0.5537036899477243],"occluded":false},{"timestamp":10.76075,"box":[0.3812500238418579,0.02777777798473835,0.6520833671092987,0.5120370257645845],"occluded":false},{"timestamp":11.011,"box":[0.3906250298023224,0.020370369777083397,0.6552083790302277,0.38703704439103603],"occluded":false},{"timestamp":11.26125,"box":[0.39270836114883423,0.02777777798473835,0.6562500298023224,0.44259257800877094],"occluded":false},{"timestamp":11.5115,"box":[0.39114585518836975,0.035185184329748154,0.6562500298023224,0.5537036918103695],"occluded":false},{"timestamp":11.76175,"box":[0.3838541805744171,0.037962961941957474,0.6427083611488342,0.5814814455807209],"occluded":false},{"timestamp":12.012,"box":[0.3734375238418579,0.03333333134651184,0.6385416984558105,0.3722222149372101],"occluded":false},{"timestamp":12.26225,"box":[0.38333335518836975,0.032407406717538834,0.6473958790302277,0.374074075371027],"occluded":false},{"timestamp":12.5125,"box":[0.3776041865348816,0.03333333134651184,0.6427083611488342,0.7898148000240326],"occluded":false},{"timestamp":12.762749999999999,"box":[0.3812500238418579,0.03333333134651184,0.6442708671092987,0.5537036955356598],"occluded":false},{"timestamp":13.013,"box":[0.3817708492279053,0.034259259700775146,0.6447916924953461,0.470370352268219],"occluded":false},{"timestamp":13.26325,"box":[0.3817708492279053,0.03611110895872116,0.6442708671092987,0.5398148223757744],"occluded":false},{"timestamp":13.513499999999999,"box":[0.3843750059604645,0.03611110895872116,0.6463541984558105,0.4842592403292656],"occluded":false},{"timestamp":13.76375,"box":[0.3817708492279053,0.035185184329748154,0.6442708671092987,0.5120370350778103],"occluded":false},{"timestamp":14.014,"box":[0.3984375298023224,0.03703703731298447,0.6656250357627869,0.609259270131588],"occluded":false},{"timestamp":14.264249999999999,"box":[0.4010416865348816,0.040740739554166794,0.6630208790302277,0.5120370201766491],"occluded":false},{"timestamp":14.5145,"box":[0.39791667461395264,0.0416666641831398,0.6567708551883698,0.6092592254281044],"occluded":false},{"timestamp":14.76475,"box":[0.3942708671092987,0.040740739554166794,0.6536458730697632,0.3870370201766491],"occluded":false},{"timestamp":15.014999999999999,"box":[0.39375001192092896,0.0416666641831398,0.6531250178813934,0.5953703299164772],"occluded":false},{"timestamp":15.26525,"box":[0.39270836114883423,0.04259259253740311,0.6526042222976685,0.525925911962986],"occluded":false},{"timestamp":15.5155,"box":[0.3916666805744171,0.0416666641831398,0.6515625417232513,0.5675925984978676],"occluded":false},{"timestamp":15.765749999999999,"box":[0.38593751192092896,0.0416666641831398,0.6458333730697632,0.5953703299164772],"occluded":false},{"timestamp":16.016,"box":[0.37968751788139343,0.04259259253740311,0.6390625238418579,0.36574072390794754],"occluded":false}],"scores":[0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322,0.859649121761322]},{"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},{"noun_phrase":"cap","object_id":0,"presentation_id":2,"observations":[{"timestamp":0.0,"box":[0.4062500298023224,0.014814814552664757,0.679166704416275,0.34722222946584225],"occluded":false},{"timestamp":0.25025,"box":[0.4270833432674408,0.014814814552664757,0.7151041924953461,0.5120370294898748],"occluded":false},{"timestamp":0.5005,"box":[0.4411458671092987,0.02222222089767456,0.736979216337204,0.3574073910713196],"occluded":false},{"timestamp":0.75075,"box":[0.4359375238418579,0.023148147389292717,0.7286458611488342,0.5537036936730146],"occluded":false},{"timestamp":1.001,"box":[0.4203125238418579,0.023148147389292717,0.7020833790302277,0.5259259026497602],"occluded":false},{"timestamp":1.25125,"box":[0.40416669845581055,0.021296296268701553,0.6703125536441803,0.5259259082376957],"occluded":false},{"timestamp":1.5015,"box":[0.39479169249534607,0.013888888992369175,0.6630208790302277,0.595370345748961],"occluded":false},{"timestamp":1.75175,"box":[0.3838541805744171,0.018518518656492233,0.6536458730697632,0.5537036769092083],"occluded":false},{"timestamp":2.002,"box":[0.3864583671092987,0.023148147389292717,0.6567708849906921,0.5675925891846418],"occluded":false},{"timestamp":2.25225,"box":[0.3958333432674408,0.01944444328546524,0.6697916984558105,0.5398148074746132],"occluded":false},{"timestamp":2.5025,"box":[0.3958333432674408,0.018518518656492233,0.6729166805744171,0.6092592589557171],"occluded":false},{"timestamp":2.75275,"box":[0.39635419845581055,0.020370369777083397,0.674479216337204,0.5537036713212729],"occluded":false},{"timestamp":3.003,"box":[0.40052086114883423,0.018518518656492233,0.6807292103767395,0.5398147813975811],"occluded":false},{"timestamp":3.25325,"box":[0.4062500298023224,0.010185184888541698,0.6848958730697632,0.5537036685273051],"occluded":false},{"timestamp":3.5035,"box":[0.40781253576278687,0.01111111044883728,0.6848958730697632,0.4564814865589142],"occluded":false},{"timestamp":3.7537499999999997,"box":[0.4088541865348816,0.015740741044282913,0.686979204416275,0.4981481321156025],"occluded":false},{"timestamp":4.004,"box":[0.4052083492279053,0.026851851493120193,0.679166704416275,0.36574073508381844],"occluded":false},{"timestamp":4.25425,"box":[0.4036458432674408,0.024074073880910873,0.6802083551883698,0.4842592440545559],"occluded":false},{"timestamp":4.5045,"box":[0.40156251192092896,0.02222222089767456,0.6802083551883698,0.470370352268219],"occluded":false},{"timestamp":4.75475,"box":[0.40000003576278687,0.024074073880910873,0.6781250536441803,0.4842592440545559],"occluded":false},{"timestamp":5.005,"box":[0.3994791805744171,0.02500000037252903,0.6744791865348816,0.5398147944360971],"occluded":false},{"timestamp":5.25525,"box":[0.3984375298023224,0.024074073880910873,0.6734375357627869,0.5398148261010647],"occluded":false},{"timestamp":5.5055,"box":[0.3973958492279053,0.024074073880910873,0.6718750298023224,0.3583333157002926],"occluded":false},{"timestamp":5.75575,"box":[0.3984375298023224,0.024074073880910873,0.6708333790302277,0.5537037216126919],"occluded":false},{"timestamp":6.006,"box":[0.37552085518836975,0.017592592164874077,0.6567708551883698,0.5814814995974302],"occluded":false},{"timestamp":6.25625,"box":[0.33020836114883423,0.023148147389292717,0.6447917222976685,0.5675925891846418],"occluded":false},{"timestamp":6.5065,"box":[0.32604169845581055,0.025925925001502037,0.6411458849906921,0.581481447443366],"occluded":false},{"timestamp":6.756749999999999,"box":[0.3395833373069763,0.024074073880910873,0.6442708671092987,0.4148147962987423],"occluded":false},{"timestamp":7.007,"box":[0.3593750298023224,0.02222222089767456,0.6427083909511566,0.539814829826355],"occluded":false},{"timestamp":7.25725,"box":[0.36770835518836975,0.02500000037252903,0.6536458730697632,0.5120370332151651],"occluded":false},{"timestamp":7.507499999999999,"box":[0.36770835518836975,0.018518518656492233,0.6661458611488342,0.5814814679324627],"occluded":false},{"timestamp":7.75775,"box":[0.3760416805744171,0.017592592164874077,0.6677083671092987,0.45648146979510784],"occluded":false},{"timestamp":8.008,"box":[0.3765625059604645,0.012037036940455437,0.6645833551883698,0.369444428011775],"occluded":false},{"timestamp":8.25825,"box":[0.3760416805744171,0.0,0.6598958671092987,0.4842592477798462],"occluded":false},{"timestamp":8.5085,"box":[0.3786458671092987,0.0,0.6619792282581329,0.4981481432914734],"occluded":false},{"timestamp":8.75875,"box":[0.3765625059604645,0.0,0.6552083492279053,0.4009259045124054],"occluded":false},{"timestamp":9.009,"box":[0.37552085518836975,0.009259259328246117,0.6500000357627869,0.5537037011235952],"occluded":false},{"timestamp":9.25925,"box":[0.37031251192092896,0.018518518656492233,0.6453125178813934,0.372222226113081],"occluded":false},{"timestamp":9.5095,"box":[0.37031251192092896,0.023148147389292717,0.6442708671092987,0.37407406978309155],"occluded":false},{"timestamp":9.75975,"box":[0.3697916865348816,0.026851851493120193,0.647916704416275,0.5675925798714161],"occluded":false},{"timestamp":10.01,"box":[0.37031251192092896,0.023148147389292717,0.6500000357627869,0.49814814142882824],"occluded":false},{"timestamp":10.26025,"box":[0.37812501192092896,0.023148147389292717,0.6567708551883698,0.5120370369404554],"occluded":false},{"timestamp":10.5105,"box":[0.37916669249534607,0.02500000037252903,0.6546875536441803,0.5537036899477243],"occluded":false},{"timestamp":10.76075,"box":[0.3812500238418579,0.02777777798473835,0.6520833671092987,0.5120370257645845],"occluded":false},{"timestamp":11.011,"box":[0.3906250298023224,0.020370369777083397,0.6552083790302277,0.38703704439103603],"occluded":false},{"timestamp":11.26125,"box":[0.39270836114883423,0.02777777798473835,0.6562500298023224,0.44259257800877094],"occluded":false},{"timestamp":11.5115,"box":[0.39114585518836975,0.035185184329748154,0.6562500298023224,0.5537036918103695],"occluded":false},{"timestamp":11.76175,"box":[0.3838541805744171,0.037962961941957474,0.6427083611488342,0.5814814455807209],"occluded":false},{"timestamp":12.012,"box":[0.3734375238418579,0.03333333134651184,0.6385416984558105,0.3722222149372101],"occluded":false},{"timestamp":12.26225,"box":[0.38333335518836975,0.032407406717538834,0.6473958790302277,0.374074075371027],"occluded":false},{"timestamp":12.5125,"box":[0.3776041865348816,0.03333333134651184,0.6427083611488342,0.7898148000240326],"occluded":false},{"timestamp":12.762749999999999,"box":[0.3812500238418579,0.03333333134651184,0.6442708671092987,0.5537036955356598],"occluded":false},{"timestamp":13.013,"box":[0.3817708492279053,0.034259259700775146,0.6447916924953461,0.470370352268219],"occluded":false},{"timestamp":13.26325,"box":[0.3817708492279053,0.03611110895872116,0.6442708671092987,0.5398148223757744],"occluded":false},{"timestamp":13.513499999999999,"box":[0.3843750059604645,0.03611110895872116,0.6463541984558105,0.4842592403292656],"occluded":false},{"timestamp":13.76375,"box":[0.3817708492279053,0.035185184329748154,0.6442708671092987,0.5120370350778103],"occluded":false},{"timestamp":14.014,"box":[0.3984375298023224,0.03703703731298447,0.6656250357627869,0.609259270131588],"occluded":false},{"timestamp":14.264249999999999,"box":[0.4010416865348816,0.040740739554166794,0.6630208790302277,0.5120370201766491],"occluded":false},{"timestamp":14.5145,"box":[0.39791667461395264,0.0416666641831398,0.6567708551883698,0.6092592254281044],"occluded":false},{"timestamp":14.76475,"box":[0.3942708671092987,0.040740739554166794,0.6536458730697632,0.3870370201766491],"occluded":false},{"timestamp":15.014999999999999,"box":[0.39375001192092896,0.0416666641831398,0.6531250178813934,0.5953703299164772],"occluded":false},{"timestamp":15.26525,"box":[0.39270836114883423,0.04259259253740311,0.6526042222976685,0.525925911962986],"occluded":false},{"timestamp":15.5155,"box":[0.3916666805744171,0.0416666641831398,0.6515625417232513,0.5675925984978676],"occluded":false},{"timestamp":15.765749999999999,"box":[0.38593751192092896,0.0416666641831398,0.6458333730697632,0.5953703299164772],"occluded":false},{"timestamp":16.016,"box":[0.37968751788139343,0.04259259253740311,0.6390625238418579,0.36574072390794754],"occluded":false}],"mean_score":0.859649121761322}],"dropped":{},"fates":[{"noun_phrase":"cap","object_id":0,"frames":65,"mean_score":0.859649121761322,"mean_area":0.13216128477206876,"outcome":"kept","presentation_id":2,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"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":2,"as_video":true,"fps":1.0,"max_frames":30},"outputs":{"media":"/home/zzk/tmp/rmot_smoke_real_86260/renders/--Dx9HVcQWs_chunk_002_scene_013_choices.mp4","media_bytes":2610411,"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","temperature":0.0,"as_video":true,"media_bytes":2610411,"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 talking and gesturing with his hands\n2. the green and black cap worn by the man\n3. the blue baseball cap","questions":["the man talking and gesturing with his hands","the green and black cap worn by the man","the blue baseball cap"]},"outputs":{"raw_response":"{\n  \"answers\": [\n    {\n      \"description_index\": 1,\n      \"matching_ids\": [1],\n      \"start_frame\": 1,\n      \"end_frame\": 17,\n      \"confidence\": 1.0\n    },\n    {\n      \"description_index\": 2,\n      \"matching_ids\": [2],\n      \"start_frame\": 1,\n      \"end_frame\": 17,\n      \"confidence\": 1.0\n    },\n    {\n      \"description_index\": 3,\n      \"matching_ids\": [],\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},{"description_index":2,"matching_ids":[2],"start_frame":1,"end_frame":17,"confidence":1.0},{"description_index":3,"matching_ids":[],"start_frame":1,"end_frame":17,"confidence":1.0}],"answers":[{"expression":"the man talking and gesturing with his hands","matching_presentation_ids":[1],"start_choice_frame":1,"end_choice_frame":17,"confidence":1.0,"self_reported_confidence":1.0},{"expression":"the green and black cap worn by the man","matching_presentation_ids":[2],"start_choice_frame":1,"end_choice_frame":17,"confidence":1.0,"self_reported_confidence":1.0},{"expression":"the blue baseball cap","matching_presentation_ids":[],"start_choice_frame":1,"end_choice_frame":17,"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 man talking and gesturing with his hands","kind":"motion","is_negative":false,"tracks":[{"timestamps":[0.0,0.25025,0.5005,0.75075,1.001,1.25125,1.5015,1.75175,2.002,2.25225,2.5025,2.75275,3.003,3.25325,3.5035,3.7537499999999997,4.004,4.25425,4.5045,4.75475,5.005,5.25525,5.5055,5.75575,6.006,6.25625,6.5065,6.756749999999999,7.007,7.25725,7.507499999999999,7.75775,8.008,8.25825,8.5085,8.75875,9.009,9.25925,9.5095,9.75975,10.01,10.26025,10.5105,10.76075,11.011,11.26125,11.5115,11.76175,12.012,12.26225,12.5125,12.762749999999999,13.013,13.26325,13.513499999999999,13.76375,14.014,14.264249999999999,14.5145,14.76475,15.014999999999999,15.26525,15.5155,15.765749999999999,16.016],"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]]}]},{"expression":"the green and black cap worn by the man","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":[[0.4062500298023224,0.014814814552664757,0.679166704416275,0.34722222946584225],[0.4270833432674408,0.014814814552664757,0.7151041924953461,0.5120370294898748],[0.4411458671092987,0.02222222089767456,0.736979216337204,0.3574073910713196],[0.4359375238418579,0.023148147389292717,0.7286458611488342,0.5537036936730146],[0.4203125238418579,0.023148147389292717,0.7020833790302277,0.5259259026497602],[0.40416669845581055,0.021296296268701553,0.6703125536441803,0.5259259082376957],[0.39479169249534607,0.013888888992369175,0.6630208790302277,0.595370345748961],[0.3838541805744171,0.018518518656492233,0.6536458730697632,0.5537036769092083],[0.3864583671092987,0.023148147389292717,0.6567708849906921,0.5675925891846418],[0.3958333432674408,0.01944444328546524,0.6697916984558105,0.5398148074746132],[0.3958333432674408,0.018518518656492233,0.6729166805744171,0.6092592589557171],[0.39635419845581055,0.020370369777083397,0.674479216337204,0.5537036713212729],[0.40052086114883423,0.018518518656492233,0.6807292103767395,0.5398147813975811],[0.4062500298023224,0.010185184888541698,0.6848958730697632,0.5537036685273051],[0.40781253576278687,0.01111111044883728,0.6848958730697632,0.4564814865589142],[0.4088541865348816,0.015740741044282913,0.686979204416275,0.4981481321156025],[0.4052083492279053,0.026851851493120193,0.679166704416275,0.36574073508381844],[0.4036458432674408,0.024074073880910873,0.6802083551883698,0.4842592440545559],[0.40156251192092896,0.02222222089767456,0.6802083551883698,0.470370352268219],[0.40000003576278687,0.024074073880910873,0.6781250536441803,0.4842592440545559],[0.3994791805744171,0.02500000037252903,0.6744791865348816,0.5398147944360971],[0.3984375298023224,0.024074073880910873,0.6734375357627869,0.5398148261010647],[0.3973958492279053,0.024074073880910873,0.6718750298023224,0.3583333157002926],[0.3984375298023224,0.024074073880910873,0.6708333790302277,0.5537037216126919],[0.37552085518836975,0.017592592164874077,0.6567708551883698,0.5814814995974302],[0.33020836114883423,0.023148147389292717,0.6447917222976685,0.5675925891846418],[0.32604169845581055,0.025925925001502037,0.6411458849906921,0.581481447443366],[0.3395833373069763,0.024074073880910873,0.6442708671092987,0.4148147962987423],[0.3593750298023224,0.02222222089767456,0.6427083909511566,0.539814829826355],[0.36770835518836975,0.02500000037252903,0.6536458730697632,0.5120370332151651],[0.36770835518836975,0.018518518656492233,0.6661458611488342,0.5814814679324627],[0.3760416805744171,0.017592592164874077,0.6677083671092987,0.45648146979510784],[0.3765625059604645,0.012037036940455437,0.6645833551883698,0.369444428011775],[0.3760416805744171,0.0,0.6598958671092987,0.4842592477798462],[0.3786458671092987,0.0,0.6619792282581329,0.4981481432914734],[0.3765625059604645,0.0,0.6552083492279053,0.4009259045124054],[0.37552085518836975,0.009259259328246117,0.6500000357627869,0.5537037011235952],[0.37031251192092896,0.018518518656492233,0.6453125178813934,0.372222226113081],[0.37031251192092896,0.023148147389292717,0.6442708671092987,0.37407406978309155],[0.3697916865348816,0.026851851493120193,0.647916704416275,0.5675925798714161],[0.37031251192092896,0.023148147389292717,0.6500000357627869,0.49814814142882824],[0.37812501192092896,0.023148147389292717,0.6567708551883698,0.5120370369404554],[0.37916669249534607,0.02500000037252903,0.6546875536441803,0.5537036899477243],[0.3812500238418579,0.02777777798473835,0.6520833671092987,0.5120370257645845],[0.3906250298023224,0.020370369777083397,0.6552083790302277,0.38703704439103603],[0.39270836114883423,0.02777777798473835,0.6562500298023224,0.44259257800877094],[0.39114585518836975,0.035185184329748154,0.6562500298023224,0.5537036918103695],[0.3838541805744171,0.037962961941957474,0.6427083611488342,0.5814814455807209],[0.3734375238418579,0.03333333134651184,0.6385416984558105,0.3722222149372101],[0.38333335518836975,0.032407406717538834,0.6473958790302277,0.374074075371027],[0.3776041865348816,0.03333333134651184,0.6427083611488342,0.7898148000240326],[0.3812500238418579,0.03333333134651184,0.6442708671092987,0.5537036955356598],[0.3817708492279053,0.034259259700775146,0.6447916924953461,0.470370352268219],[0.3817708492279053,0.03611110895872116,0.6442708671092987,0.5398148223757744],[0.3843750059604645,0.03611110895872116,0.6463541984558105,0.4842592403292656],[0.3817708492279053,0.035185184329748154,0.6442708671092987,0.5120370350778103],[0.3984375298023224,0.03703703731298447,0.6656250357627869,0.609259270131588],[0.4010416865348816,0.040740739554166794,0.6630208790302277,0.5120370201766491],[0.39791667461395264,0.0416666641831398,0.6567708551883698,0.6092592254281044],[0.3942708671092987,0.040740739554166794,0.6536458730697632,0.3870370201766491],[0.39375001192092896,0.0416666641831398,0.6531250178813934,0.5953703299164772],[0.39270836114883423,0.04259259253740311,0.6526042222976685,0.525925911962986],[0.3916666805744171,0.0416666641831398,0.6515625417232513,0.5675925984978676],[0.38593751192092896,0.0416666641831398,0.6458333730697632,0.5953703299164772],[0.37968751788139343,0.04259259253740311,0.6390625238418579,0.36574072390794754]]}]},{"expression":"the blue baseball cap","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/--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 talking and gesturing with his hands","kind":"motion","is_negative":false,"tracks":[{"timestamps":[0.0,0.25025,0.5005,0.75075,1.001,1.25125,1.5015,1.75175,2.002,2.25225,2.5025,2.75275,3.003,3.25325,3.5035,3.7537499999999997,4.004,4.25425,4.5045,4.75475,5.005,5.25525,5.5055,5.75575,6.006,6.25625,6.5065,6.756749999999999,7.007,7.25725,7.507499999999999,7.75775,8.008,8.25825,8.5085,8.75875,9.009,9.25925,9.5095,9.75975,10.01,10.26025,10.5105,10.76075,11.011,11.26125,11.5115,11.76175,12.012,12.26225,12.5125,12.762749999999999,13.013,13.26325,13.513499999999999,13.76375,14.014,14.264249999999999,14.5145,14.76475,15.014999999999999,15.26525,15.5155,15.765749999999999,16.016],"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]]}]},{"expression":"the green and black cap worn by the man","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":[[0.4062500298023224,0.014814814552664757,0.679166704416275,0.34722222946584225],[0.4270833432674408,0.014814814552664757,0.7151041924953461,0.5120370294898748],[0.4411458671092987,0.02222222089767456,0.736979216337204,0.3574073910713196],[0.4359375238418579,0.023148147389292717,0.7286458611488342,0.5537036936730146],[0.4203125238418579,0.023148147389292717,0.7020833790302277,0.5259259026497602],[0.40416669845581055,0.021296296268701553,0.6703125536441803,0.5259259082376957],[0.39479169249534607,0.013888888992369175,0.6630208790302277,0.595370345748961],[0.3838541805744171,0.018518518656492233,0.6536458730697632,0.5537036769092083],[0.3864583671092987,0.023148147389292717,0.6567708849906921,0.5675925891846418],[0.3958333432674408,0.01944444328546524,0.6697916984558105,0.5398148074746132],[0.3958333432674408,0.018518518656492233,0.6729166805744171,0.6092592589557171],[0.39635419845581055,0.020370369777083397,0.674479216337204,0.5537036713212729],[0.40052086114883423,0.018518518656492233,0.6807292103767395,0.5398147813975811],[0.4062500298023224,0.010185184888541698,0.6848958730697632,0.5537036685273051],[0.40781253576278687,0.01111111044883728,0.6848958730697632,0.4564814865589142],[0.4088541865348816,0.015740741044282913,0.686979204416275,0.4981481321156025],[0.4052083492279053,0.026851851493120193,0.679166704416275,0.36574073508381844],[0.4036458432674408,0.024074073880910873,0.6802083551883698,0.4842592440545559],[0.40156251192092896,0.02222222089767456,0.6802083551883698,0.470370352268219],[0.40000003576278687,0.024074073880910873,0.6781250536441803,0.4842592440545559],[0.3994791805744171,0.02500000037252903,0.6744791865348816,0.5398147944360971],[0.3984375298023224,0.024074073880910873,0.6734375357627869,0.5398148261010647],[0.3973958492279053,0.024074073880910873,0.6718750298023224,0.3583333157002926],[0.3984375298023224,0.024074073880910873,0.6708333790302277,0.5537037216126919],[0.37552085518836975,0.017592592164874077,0.6567708551883698,0.5814814995974302],[0.33020836114883423,0.023148147389292717,0.6447917222976685,0.5675925891846418],[0.32604169845581055,0.025925925001502037,0.6411458849906921,0.581481447443366],[0.3395833373069763,0.024074073880910873,0.6442708671092987,0.4148147962987423],[0.3593750298023224,0.02222222089767456,0.6427083909511566,0.539814829826355],[0.36770835518836975,0.02500000037252903,0.6536458730697632,0.5120370332151651],[0.36770835518836975,0.018518518656492233,0.6661458611488342,0.5814814679324627],[0.3760416805744171,0.017592592164874077,0.6677083671092987,0.45648146979510784],[0.3765625059604645,0.012037036940455437,0.6645833551883698,0.369444428011775],[0.3760416805744171,0.0,0.6598958671092987,0.4842592477798462],[0.3786458671092987,0.0,0.6619792282581329,0.4981481432914734],[0.3765625059604645,0.0,0.6552083492279053,0.4009259045124054],[0.37552085518836975,0.009259259328246117,0.6500000357627869,0.5537037011235952],[0.37031251192092896,0.018518518656492233,0.6453125178813934,0.372222226113081],[0.37031251192092896,0.023148147389292717,0.6442708671092987,0.37407406978309155],[0.3697916865348816,0.026851851493120193,0.647916704416275,0.5675925798714161],[0.37031251192092896,0.023148147389292717,0.6500000357627869,0.49814814142882824],[0.37812501192092896,0.023148147389292717,0.6567708551883698,0.5120370369404554],[0.37916669249534607,0.02500000037252903,0.6546875536441803,0.5537036899477243],[0.3812500238418579,0.02777777798473835,0.6520833671092987,0.5120370257645845],[0.3906250298023224,0.020370369777083397,0.6552083790302277,0.38703704439103603],[0.39270836114883423,0.02777777798473835,0.6562500298023224,0.44259257800877094],[0.39114585518836975,0.035185184329748154,0.6562500298023224,0.5537036918103695],[0.3838541805744171,0.037962961941957474,0.6427083611488342,0.5814814455807209],[0.3734375238418579,0.03333333134651184,0.6385416984558105,0.3722222149372101],[0.38333335518836975,0.032407406717538834,0.6473958790302277,0.374074075371027],[0.3776041865348816,0.03333333134651184,0.6427083611488342,0.7898148000240326],[0.3812500238418579,0.03333333134651184,0.6442708671092987,0.5537036955356598],[0.3817708492279053,0.034259259700775146,0.6447916924953461,0.470370352268219],[0.3817708492279053,0.03611110895872116,0.6442708671092987,0.5398148223757744],[0.3843750059604645,0.03611110895872116,0.6463541984558105,0.4842592403292656],[0.3817708492279053,0.035185184329748154,0.6442708671092987,0.5120370350778103],[0.3984375298023224,0.03703703731298447,0.6656250357627869,0.609259270131588],[0.4010416865348816,0.040740739554166794,0.6630208790302277,0.5120370201766491],[0.39791667461395264,0.0416666641831398,0.6567708551883698,0.6092592254281044],[0.3942708671092987,0.040740739554166794,0.6536458730697632,0.3870370201766491],[0.39375001192092896,0.0416666641831398,0.6531250178813934,0.5953703299164772],[0.39270836114883423,0.04259259253740311,0.6526042222976685,0.525925911962986],[0.3916666805744171,0.0416666641831398,0.6515625417232513,0.5675925984978676],[0.38593751192092896,0.0416666641831398,0.6458333730697632,0.5953703299164772],[0.37968751788139343,0.04259259253740311,0.6390625238418579,0.36574072390794754]]}]},{"expression":"the blue baseball cap","kind":"appearance","is_negative":true,"tracks":[]}],"rmot_expression_count":3,"rmot_negative_count":1,"rmot_motion_count":1}}}}},{"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","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 cyclist who drifts furthest back on the right side of the road\",\n      \"noun_phrase\": \"cyclist in pink jersey\",\n      \"kind\": \"motion\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the cyclist with a white helmet who rides behind the leader on the left\",\n      \"noun_phrase\": \"cyclist with white helmet\",\n      \"kind\": \"spatial\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the cyclist wearing a green helmet\",\n      \"noun_phrase\": \"cyclist with green helmet\",\n      \"kind\": \"appearance\",\n      \"expected_count\": 0,\n      \"is_negative\": true\n    }\n  ]\n}","raw_items":[{"expression":"the cyclist who drifts furthest back on the right side of the road","noun_phrase":"cyclist in pink jersey","kind":"motion","expected_count":1,"is_negative":false},{"expression":"the cyclist with a white helmet who rides behind the leader on the left","noun_phrase":"cyclist with white helmet","kind":"spatial","expected_count":1,"is_negative":false},{"expression":"the cyclist wearing a green helmet","noun_phrase":"cyclist with green helmet","kind":"appearance","expected_count":0,"is_negative":true}],"accepted":[{"expression":"the cyclist who drifts furthest back on the right side of the road","noun_phrase":"cyclist in pink jersey","kind":"motion","expected_count":1,"is_negative":false},{"expression":"the cyclist with a white helmet who rides behind the leader on the left","noun_phrase":"cyclist with white helmet","kind":"spatial","expected_count":1,"is_negative":false},{"expression":"the cyclist wearing a green helmet","noun_phrase":"cyclist with green helmet","kind":"appearance","expected_count":0,"is_negative":true}],"gated":[]}},"track":{"inputs":{"tracker":"Sam31MultiplexTracker","model_id":"/mnt/ceph3/ec/jacky/sam3.1","noun_phrases":["cyclist in pink jersey","cyclist with green helmet","cyclist with white helmet"],"expressions_in":3,"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":3},"outputs":{"raw_tracks":[{"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]},{"noun_phrase":"cyclist with white helmet","object_id":0,"observations":[{"timestamp":0.0,"box":[0.5869792103767395,0.3722222149372101,0.6781250461935997,0.6888888776302338],"occluded":false},{"timestamp":0.25025,"box":[0.5984375476837158,0.36574074625968933,0.6802083849906921,0.6833333373069763],"occluded":false},{"timestamp":0.5005,"box":[0.5640625357627869,0.37037035822868347,0.6786458715796471,0.6657407283782959],"occluded":false},{"timestamp":0.75075,"box":[0.5963541865348816,0.3759259283542633,0.6750000268220901,0.6574074029922485],"occluded":false},{"timestamp":1.001,"box":[0.5937500596046448,0.3768518567085266,0.668750062584877,0.6657407283782959],"occluded":false},{"timestamp":1.25125,"box":[0.5864583849906921,0.3861111104488373,0.6583333909511566,0.5537036955356598],"occluded":false},{"timestamp":1.5015,"box":[0.5843750238418579,0.39351850748062134,0.65416669100523,0.6472222208976746],"occluded":false},{"timestamp":1.75175,"box":[0.5671875476837158,0.39444443583488464,0.635416716337204,0.6611110866069794],"occluded":false},{"timestamp":2.002,"box":[0.46302086114883423,0.33796295523643494,0.6312500387430191,0.6712962687015533],"occluded":false},{"timestamp":2.25225,"box":[0.47708335518836975,0.35185185074806213,0.6260416954755783,0.6509259343147278],"occluded":false},{"timestamp":2.5025,"box":[0.5593750476837158,0.4037036895751953,0.6218750476837158,0.64907406270504],"occluded":false},{"timestamp":2.75275,"box":[0.5598958730697632,0.4037036895751953,0.6171875409781933,0.6601851582527161],"occluded":false},{"timestamp":3.003,"box":[0.5593750476837158,0.4055555462837219,0.6098958849906921,0.6499999910593033],"occluded":false},{"timestamp":3.25325,"box":[0.5505208373069763,0.412962943315506,0.6072916723787785,0.6398147940635681],"occluded":false},{"timestamp":3.5035,"box":[0.565625011920929,0.4435185194015503,0.603645846247673,0.6129629611968994],"occluded":false},{"timestamp":3.7537499999999997,"box":[0.5640625357627869,0.5435184836387634,0.5750000365078449,0.6527777388691902],"occluded":false},{"timestamp":4.004,"box":[0.5666667222976685,0.4722222089767456,0.5822917222976685,0.587037019431591],"occluded":false},{"timestamp":4.25425,"box":[0.5588542222976685,0.4722222089767456,0.5848958902060986,0.6249999850988388],"occluded":false},{"timestamp":4.5045,"box":[0.5463541746139526,0.49166664481163025,0.5864583440124989,0.6509259045124054],"occluded":false},{"timestamp":4.75475,"box":[0.5390625,0.44722220301628113,0.5864583365619183,0.6518518328666687],"occluded":false},{"timestamp":5.005,"box":[0.5432292222976685,0.4314814805984497,0.5864583924412727,0.635185182094574],"occluded":false},{"timestamp":5.25525,"box":[0.5479167103767395,0.4277777671813965,0.5703125447034836,0.5703703612089157],"occluded":false},{"timestamp":5.5055,"box":[0.5567708611488342,0.42499998211860657,0.5786458626389503,0.5935184955596924],"occluded":false},{"timestamp":5.75575,"box":[0.5770833492279053,0.45092591643333435,0.5968750175088644,0.5296296179294586],"occluded":false},{"timestamp":6.25625,"box":[0.5901042222976685,0.5046296119689941,0.6031250562518835,0.5407407209277153],"occluded":false},{"timestamp":7.25725,"box":[0.6015625596046448,0.43888887763023376,0.6260417271405458,0.527777761220932],"occluded":false},{"timestamp":7.507499999999999,"box":[0.5328125357627869,0.4166666567325592,0.6072917059063911,0.5268518403172493],"occluded":false},{"timestamp":7.75775,"box":[0.5671875476837158,0.4175925850868225,0.59218754991889,0.6444444358348846],"occluded":false},{"timestamp":8.008,"box":[0.5552083849906921,0.4185185134410858,0.5750000532716513,0.6231481432914734],"occluded":false},{"timestamp":8.25825,"box":[0.542187511920929,0.4203703701496124,0.5921875163912773,0.6537037044763565],"occluded":false},{"timestamp":8.75875,"box":[0.5244792103767395,0.4287036955356598,0.5718750469386578,0.6351851671934128],"occluded":false},{"timestamp":9.009,"box":[0.5057291984558105,0.4259259104728699,0.5552083663642406,0.6518518328666687],"occluded":false},{"timestamp":9.25925,"box":[0.5062500238418579,0.4259259104728699,0.5572916939854622,0.6537036895751953],"occluded":false},{"timestamp":9.5095,"box":[0.4911458492279053,0.4296296238899231,0.5536458492279053,0.635185182094574],"occluded":false},{"timestamp":9.75975,"box":[0.5,0.4296296238899231,0.5484375022351742,0.6398148089647293],"occluded":false},{"timestamp":10.01,"box":[0.5005208849906921,0.4296296238899231,0.5489583872258663,0.64907406270504],"occluded":false},{"timestamp":10.26025,"box":[0.483854204416275,0.43611109256744385,0.5317708738148212,0.6416666507720947],"occluded":false},{"timestamp":10.5105,"box":[0.4843750298023224,0.43888887763023376,0.5307292006909847,0.6342592388391495],"occluded":false},{"timestamp":10.76075,"box":[0.5,0.43611109256744385,0.5401041693985462,0.6407407224178314],"occluded":false},{"timestamp":11.011,"box":[0.5036458373069763,0.43981480598449707,0.5401041731238365,0.6370370239019394],"occluded":false},{"timestamp":11.26125,"box":[0.5041667222976685,0.43888887763023376,0.5395833924412727,0.6212962865829468],"occluded":false},{"timestamp":11.5115,"box":[0.5140625238418579,0.4435185194015503,0.5375000257045031,0.635185182094574],"occluded":false},{"timestamp":11.76175,"box":[0.5083333849906921,0.4453703761100769,0.5401042215526104,0.6296296268701553],"occluded":false},{"timestamp":12.012,"box":[0.5135416984558105,0.44907405972480774,0.5406250339001417,0.6222222000360489],"occluded":false},{"timestamp":12.26225,"box":[0.5026041865348816,0.44999998807907104,0.5364583544433117,0.6222222000360489],"occluded":false},{"timestamp":12.5125,"box":[0.499479204416275,0.44814813137054443,0.5328125394880772,0.6277777552604675],"occluded":false},{"timestamp":12.762749999999999,"box":[0.4911458492279053,0.4287036955356598,0.5234375186264515,0.6185185015201569],"occluded":false},{"timestamp":13.013,"box":[0.4911458492279053,0.45185184478759766,0.5213541835546494,0.6166666597127914],"occluded":false},{"timestamp":13.26325,"box":[0.49687501788139343,0.45370370149612427,0.5296875201165676,0.6240740716457367],"occluded":false},{"timestamp":13.513499999999999,"box":[0.4989583492279053,0.4546296298503876,0.5260416846722364,0.6175925880670547],"occluded":false},{"timestamp":13.76375,"box":[0.49635419249534607,0.46388888359069824,0.5239583607763052,0.6064814776182175],"occluded":false},{"timestamp":14.014,"box":[0.4906250238418579,0.460185170173645,0.515104191377759,0.6157407164573669],"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]},{"noun_phrase":"cyclist with white helmet","object_id":1,"observations":[{"timestamp":0.0,"box":[0.4385416805744171,0.412962943315506,0.4802083484828472,0.6861110925674438],"occluded":false},{"timestamp":0.25025,"box":[0.4385416805744171,0.4046296179294586,0.4911458492279053,0.6833333075046539],"occluded":false},{"timestamp":0.5005,"box":[0.42760419845581055,0.4000000059604645,0.48750003427267075,0.6842592656612396],"occluded":false},{"timestamp":0.75075,"box":[0.41822919249534607,0.4055555462837219,0.4776041954755783,0.6712962687015533],"occluded":false},{"timestamp":1.001,"box":[0.41041669249534607,0.4037036895751953,0.4635416939854622,0.6888888776302338],"occluded":false},{"timestamp":1.25125,"box":[0.40260419249534607,0.4037036895751953,0.4682291969656944,0.6925925612449646],"occluded":false},{"timestamp":1.5015,"box":[0.40000003576278687,0.4009259045124054,0.4697917029261589,0.6611110866069794],"occluded":false},{"timestamp":1.75175,"box":[0.39010417461395264,0.4037036895751953,0.45000001043081284,0.6601851582527161],"occluded":false},{"timestamp":2.002,"box":[0.39010417461395264,0.40740740299224854,0.45000001043081284,0.6527777761220932],"occluded":false},{"timestamp":2.25225,"box":[0.3958333432674408,0.41111111640930176,0.4567708447575569,0.6842592656612396],"occluded":false},{"timestamp":2.5025,"box":[0.40416669845581055,0.4175925850868225,0.4567708671092987,0.6574074029922485],"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":[0.4203125238418579,0.42407405376434326,0.4713541939854622,0.6685184985399246],"occluded":false},{"timestamp":3.5035,"box":[0.42760419845581055,0.42407405376434326,0.48072919994592667,0.6527777463197708],"occluded":false},{"timestamp":3.7537499999999997,"box":[0.42500001192092896,0.4268518388271332,0.6020833551883698,0.6685184985399246],"occluded":false},{"timestamp":4.004,"box":[0.43802085518836975,0.42499998211860657,0.5744791924953461,0.6675925701856613],"occluded":false},{"timestamp":4.25425,"box":[0.44218751788139343,0.4268518388271332,0.49010418727993965,0.6509259045124054],"occluded":false},{"timestamp":4.5045,"box":[0.4375000298023224,0.4287036955356598,0.4614583645015955,0.6259259134531021],"occluded":false},{"timestamp":4.75475,"box":[0.444791704416275,0.4416666626930237,0.47447920590639114,0.6703703552484512],"occluded":false},{"timestamp":5.005,"box":[0.45104169845581055,0.460185170173645,0.47708336636424065,0.6398147940635681],"occluded":false},{"timestamp":5.25525,"box":[0.4515625238418579,0.4166666567325592,0.479687524959445,0.6416666507720947],"occluded":false},{"timestamp":5.5055,"box":[0.45468753576278687,0.41018518805503845,0.482291704043746,0.6805555522441864],"occluded":false},{"timestamp":5.75575,"box":[0.45625001192092896,0.40648147463798523,0.5078125149011612,0.6787036955356598],"occluded":false},{"timestamp":6.006,"box":[0.46302086114883423,0.39814814925193787,0.602083370089531,0.6638888716697693],"occluded":false},{"timestamp":6.25625,"box":[0.46666669845581055,0.39444443583488464,0.5302083715796471,0.6749999821186066],"occluded":false},{"timestamp":6.5065,"box":[0.47187501192092896,0.39537036418914795,0.6161458492279053,0.6777777671813965],"occluded":false},{"timestamp":6.756749999999999,"box":[0.47343751788139343,0.39351850748062134,0.6026041954755783,0.6851851642131805],"occluded":false},{"timestamp":7.007,"box":[0.46302086114883423,0.39259257912635803,0.6437500417232513,0.6833333075046539],"occluded":false},{"timestamp":7.25725,"box":[0.4619791805744171,0.3916666507720947,0.5187500156462193,0.6638888716697693],"occluded":false},{"timestamp":7.507499999999999,"box":[0.4520833492279053,0.3907407224178314,0.5333333536982536,0.6898148059844971],"occluded":false},{"timestamp":7.75775,"box":[0.44062501192092896,0.44814813137054443,0.4723958484828472,0.6361110955476761],"occluded":false},{"timestamp":8.008,"box":[0.43385419249534607,0.46759259700775146,0.46406252682209015,0.6833333373069763],"occluded":false},{"timestamp":8.25825,"box":[0.4479166865348816,0.550000011920929,0.46197918709367514,0.6157407537102699],"occluded":false},{"timestamp":8.5085,"box":[0.40052086114883423,0.54537034034729,0.43177086114883423,0.6249999701976776],"occluded":false},{"timestamp":9.009,"box":[0.3932291865348816,0.47592592239379883,0.4088541865348816,0.5722222179174423],"occluded":false},{"timestamp":9.75975,"box":[0.3812500238418579,0.46203702688217163,0.4791666939854622,0.7259259223937988],"occluded":false},{"timestamp":10.01,"box":[0.46302086114883423,0.45092591643333435,0.4875000286847353,0.6703703552484512],"occluded":false},{"timestamp":10.26025,"box":[0.4453125298023224,0.4453703761100769,0.47968753054738045,0.6388888955116272],"occluded":false},{"timestamp":10.5105,"box":[0.36927086114883423,0.4546296298503876,0.48489587008953094,0.6888888776302338],"occluded":false},{"timestamp":10.76075,"box":[0.4755208492279053,0.44999998807907104,0.5020833499729633,0.6416666507720947],"occluded":false},{"timestamp":11.011,"box":[0.4791666865348816,0.45092591643333435,0.5083333551883698,0.6509259194135666],"occluded":false},{"timestamp":11.26125,"box":[0.3932291865348816,0.43518516421318054,0.5140625238418579,0.6712962687015533],"occluded":false},{"timestamp":11.5115,"box":[0.4723958671092987,0.4444444477558136,0.5177083685994148,0.6805555522441864],"occluded":false},{"timestamp":11.76175,"box":[0.4723958671092987,0.4018518328666687,0.516145870089531,0.6740740537643433],"occluded":false},{"timestamp":12.012,"box":[0.4755208492279053,0.4435185194015503,0.5223958529531956,0.6546296328306198],"occluded":false},{"timestamp":12.26225,"box":[0.4598958492279053,0.4611110985279083,0.5072916857898235,0.6620370149612427],"occluded":false},{"timestamp":12.5125,"box":[0.48125001788139343,0.5314814448356628,0.49843751825392246,0.6694444119930267],"occluded":false},{"timestamp":12.762749999999999,"box":[0.3932291865348816,0.24722221493721008,0.5223958641290665,0.6370370090007782],"occluded":false},{"timestamp":13.013,"box":[0.47187501192092896,0.4138888716697693,0.494791679084301,0.6444444209337234],"occluded":false},{"timestamp":13.26325,"box":[0.48854169249534607,0.44907405972480774,0.5057291928678751,0.6546296179294586],"occluded":false},{"timestamp":13.513499999999999,"box":[0.483854204416275,0.44907405972480774,0.5015625394880772,0.6148148030042648],"occluded":false},{"timestamp":13.76375,"box":[0.47968751192092896,0.42314815521240234,0.5046875141561031,0.6574074029922485],"occluded":false},{"timestamp":14.014,"box":[0.47343751788139343,0.42222222685813904,0.4979166854172945,0.631481483578682],"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]},{"noun_phrase":"cyclist with white helmet","object_id":2,"observations":[{"timestamp":1.5015,"box":[0.444791704416275,0.2370370328426361,0.5750000476837158,0.8046295940876007],"occluded":false},{"timestamp":1.75175,"box":[0.42604169249534607,0.23888888955116272,0.555208370089531,0.817592591047287],"occluded":false},{"timestamp":2.002,"box":[0.4244791865348816,0.24074073135852814,0.5614583641290665,0.7861110717058182],"occluded":false},{"timestamp":2.25225,"box":[0.42500001192092896,0.2490740716457367,0.5635416805744171,0.7750000059604645],"occluded":false},{"timestamp":2.5025,"box":[0.43125003576278687,0.2537037134170532,0.563020870089531,0.8046296238899231],"occluded":false},{"timestamp":2.75275,"box":[0.444791704416275,0.2888888716697693,0.569791704416275,0.7981480956077576],"occluded":false},{"timestamp":3.003,"box":[0.4489583671092987,0.28703704476356506,0.5723958760499954,0.7583333253860474],"occluded":false},{"timestamp":3.25325,"box":[0.4661458432674408,0.25740739703178406,0.5927083492279053,0.7722221910953522],"occluded":false},{"timestamp":3.5035,"box":[0.47864586114883423,0.2620370388031006,0.6072916984558105,0.7935184836387634],"occluded":false},{"timestamp":3.7537499999999997,"box":[0.4739583432674408,0.2638888955116272,0.6020833551883698,0.7833333015441895],"occluded":false},{"timestamp":4.004,"box":[0.47968751192092896,0.2527777850627899,0.6093750149011612,0.7537036836147308],"occluded":false},{"timestamp":4.25425,"box":[0.4713541865348816,0.2509259283542633,0.6046875268220901,0.7898148000240326],"occluded":false},{"timestamp":4.5045,"box":[0.44218751788139343,0.24444444477558136,0.5812500268220901,0.8138888627290726],"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":[0.4515625238418579,0.2212962955236435,0.6015625298023224,0.8148147910833359],"occluded":false},{"timestamp":5.5055,"box":[0.46927085518836975,0.2231481373310089,0.617708370089531,0.8259259164333344],"occluded":false},{"timestamp":5.75575,"box":[0.49427086114883423,0.21666666865348816,0.6427083760499954,0.7870370447635651],"occluded":false},{"timestamp":6.006,"box":[0.5250000357627869,0.20555555820465088,0.6781250387430191,0.8074073791503906],"occluded":false},{"timestamp":6.25625,"box":[0.546875,0.20370370149612427,0.7067708373069763,0.8222222328186035],"occluded":false},{"timestamp":6.5065,"box":[0.5546875,0.2092592567205429,0.7156250029802322,0.783333346247673],"occluded":false},{"timestamp":6.756749999999999,"box":[0.5463541746139526,0.20277777314186096,0.7161458432674408,0.7833333313465118],"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":[0.4869791865348816,0.19074073433876038,0.645833358168602,0.7842592298984528],"occluded":false},{"timestamp":7.75775,"box":[0.460416704416275,0.32222220301628113,0.6250000447034836,0.8259259164333344],"occluded":false},{"timestamp":8.008,"box":[0.43906253576278687,0.32499998807907104,0.6062500476837158,0.845370352268219],"occluded":false},{"timestamp":8.25825,"box":[0.4130208492279053,0.3574073910713196,0.5854166895151138,0.8240740597248077],"occluded":false},{"timestamp":8.5085,"box":[0.39635419845581055,0.10092592239379883,0.6369792073965073,0.8240740299224854],"occluded":false},{"timestamp":8.75875,"box":[0.3515625298023224,0.18611110746860504,0.538541704416275,0.8527777343988419],"occluded":false},{"timestamp":9.009,"box":[0.34166669845581055,0.192592591047287,0.5255208760499954,0.8398148119449615],"occluded":false},{"timestamp":9.25925,"box":[0.3296875059604645,0.1787036955356598,0.5166666805744171,0.807407408952713],"occluded":false},{"timestamp":9.5095,"box":[0.29635417461395264,0.1787036955356598,0.5223958492279053,0.8425925672054291],"occluded":false},{"timestamp":9.75975,"box":[0.30520835518836975,0.12222222238779068,0.49791669845581055,0.8453703299164772],"occluded":false},{"timestamp":10.01,"box":[0.3036458492279053,0.20000000298023224,0.489583358168602,0.8185185343027115],"occluded":false},{"timestamp":10.26025,"box":[0.2796875238418579,0.19722221791744232,0.49375003576278687,0.8185185045003891],"occluded":false},{"timestamp":10.5105,"box":[0.29635417461395264,0.20277777314186096,0.46406251192092896,0.8444444239139557],"occluded":false},{"timestamp":10.76075,"box":[0.31822916865348816,0.2074074000120163,0.49427084624767303,0.8444444388151169],"occluded":false},{"timestamp":11.011,"box":[0.3218750059604645,0.20185184478759766,0.4828125089406967,0.7990740537643433],"occluded":false},{"timestamp":11.26125,"box":[0.3135416805744171,0.19814814627170563,0.48906251788139343,0.8296296149492264],"occluded":false},{"timestamp":11.5115,"box":[0.30781251192092896,0.20555555820465088,0.49427086114883423,0.8490740656852722],"occluded":false},{"timestamp":11.76175,"box":[0.3098958432674408,0.19629628956317902,0.4921875149011612,0.8203703314065933],"occluded":false},{"timestamp":12.012,"box":[0.30677086114883423,0.20555555820465088,0.47708337008953094,0.8185185194015503],"occluded":false},{"timestamp":12.26225,"box":[0.32447919249534607,0.18518517911434174,0.5052083730697632,0.8398148268461227],"occluded":false},{"timestamp":12.5125,"box":[0.3229166865348816,0.18518517911434174,0.49947918951511383,0.7499999850988388],"occluded":false},{"timestamp":12.762749999999999,"box":[0.3229166865348816,0.13611111044883728,0.4921875298023224,0.7907407581806183],"occluded":false},{"timestamp":13.013,"box":[0.31718751788139343,0.1953703612089157,0.47864586114883423,0.8268518298864365],"occluded":false},{"timestamp":13.26325,"box":[0.36666667461395264,0.2074074000120163,0.502083346247673,0.8277777880430222],"occluded":false},{"timestamp":13.513499999999999,"box":[0.33020836114883423,0.21759258210659027,0.48750004172325134,0.8046296089887619],"occluded":false},{"timestamp":13.76375,"box":[0.3307291865348816,0.2203703671693802,0.4812500327825546,0.7888888865709305],"occluded":false},{"timestamp":14.014,"box":[0.33229169249534607,0.22962962090969086,0.476041704416275,0.7999999970197678],"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]}],"tracklets":[{"noun_phrase":"cyclist with white helmet","object_id":1,"presentation_id":1,"observations":[{"timestamp":0.0,"box":[0.4385416805744171,0.412962943315506,0.4802083484828472,0.6861110925674438],"occluded":false},{"timestamp":0.25025,"box":[0.4385416805744171,0.4046296179294586,0.4911458492279053,0.6833333075046539],"occluded":false},{"timestamp":0.5005,"box":[0.42760419845581055,0.4000000059604645,0.48750003427267075,0.6842592656612396],"occluded":false},{"timestamp":0.75075,"box":[0.41822919249534607,0.4055555462837219,0.4776041954755783,0.6712962687015533],"occluded":false},{"timestamp":1.001,"box":[0.41041669249534607,0.4037036895751953,0.4635416939854622,0.6888888776302338],"occluded":false},{"timestamp":1.25125,"box":[0.40260419249534607,0.4037036895751953,0.4682291969656944,0.6925925612449646],"occluded":false},{"timestamp":1.5015,"box":[0.40000003576278687,0.4009259045124054,0.4697917029261589,0.6611110866069794],"occluded":false},{"timestamp":1.75175,"box":[0.39010417461395264,0.4037036895751953,0.45000001043081284,0.6601851582527161],"occluded":false},{"timestamp":2.002,"box":[0.39010417461395264,0.40740740299224854,0.45000001043081284,0.6527777761220932],"occluded":false},{"timestamp":2.25225,"box":[0.3958333432674408,0.41111111640930176,0.4567708447575569,0.6842592656612396],"occluded":false},{"timestamp":2.5025,"box":[0.40416669845581055,0.4175925850868225,0.4567708671092987,0.6574074029922485],"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":[0.4203125238418579,0.42407405376434326,0.4713541939854622,0.6685184985399246],"occluded":false},{"timestamp":3.5035,"box":[0.42760419845581055,0.42407405376434326,0.48072919994592667,0.6527777463197708],"occluded":false},{"timestamp":3.7537499999999997,"box":[0.42500001192092896,0.4268518388271332,0.6020833551883698,0.6685184985399246],"occluded":false},{"timestamp":4.004,"box":[0.43802085518836975,0.42499998211860657,0.5744791924953461,0.6675925701856613],"occluded":false},{"timestamp":4.25425,"box":[0.44218751788139343,0.4268518388271332,0.49010418727993965,0.6509259045124054],"occluded":false},{"timestamp":4.5045,"box":[0.4375000298023224,0.4287036955356598,0.4614583645015955,0.6259259134531021],"occluded":false},{"timestamp":4.75475,"box":[0.444791704416275,0.4416666626930237,0.47447920590639114,0.6703703552484512],"occluded":false},{"timestamp":5.005,"box":[0.45104169845581055,0.460185170173645,0.47708336636424065,0.6398147940635681],"occluded":false},{"timestamp":5.25525,"box":[0.4515625238418579,0.4166666567325592,0.479687524959445,0.6416666507720947],"occluded":false},{"timestamp":5.5055,"box":[0.45468753576278687,0.41018518805503845,0.482291704043746,0.6805555522441864],"occluded":false},{"timestamp":5.75575,"box":[0.45625001192092896,0.40648147463798523,0.5078125149011612,0.6787036955356598],"occluded":false},{"timestamp":6.006,"box":[0.46302086114883423,0.39814814925193787,0.602083370089531,0.6638888716697693],"occluded":false},{"timestamp":6.25625,"box":[0.46666669845581055,0.39444443583488464,0.5302083715796471,0.6749999821186066],"occluded":false},{"timestamp":6.5065,"box":[0.47187501192092896,0.39537036418914795,0.6161458492279053,0.6777777671813965],"occluded":false},{"timestamp":6.756749999999999,"box":[0.47343751788139343,0.39351850748062134,0.6026041954755783,0.6851851642131805],"occluded":false},{"timestamp":7.007,"box":[0.46302086114883423,0.39259257912635803,0.6437500417232513,0.6833333075046539],"occluded":false},{"timestamp":7.25725,"box":[0.4619791805744171,0.3916666507720947,0.5187500156462193,0.6638888716697693],"occluded":false},{"timestamp":7.507499999999999,"box":[0.4520833492279053,0.3907407224178314,0.5333333536982536,0.6898148059844971],"occluded":false},{"timestamp":7.75775,"box":[0.44062501192092896,0.44814813137054443,0.4723958484828472,0.6361110955476761],"occluded":false},{"timestamp":8.008,"box":[0.43385419249534607,0.46759259700775146,0.46406252682209015,0.6833333373069763],"occluded":false},{"timestamp":8.25825,"box":[0.4479166865348816,0.550000011920929,0.46197918709367514,0.6157407537102699],"occluded":false},{"timestamp":8.5085,"box":[0.40052086114883423,0.54537034034729,0.43177086114883423,0.6249999701976776],"occluded":false},{"timestamp":9.009,"box":[0.3932291865348816,0.47592592239379883,0.4088541865348816,0.5722222179174423],"occluded":false},{"timestamp":9.75975,"box":[0.3812500238418579,0.46203702688217163,0.4791666939854622,0.7259259223937988],"occluded":false},{"timestamp":10.01,"box":[0.46302086114883423,0.45092591643333435,0.4875000286847353,0.6703703552484512],"occluded":false},{"timestamp":10.26025,"box":[0.4453125298023224,0.4453703761100769,0.47968753054738045,0.6388888955116272],"occluded":false},{"timestamp":10.5105,"box":[0.36927086114883423,0.4546296298503876,0.48489587008953094,0.6888888776302338],"occluded":false},{"timestamp":10.76075,"box":[0.4755208492279053,0.44999998807907104,0.5020833499729633,0.6416666507720947],"occluded":false},{"timestamp":11.011,"box":[0.4791666865348816,0.45092591643333435,0.5083333551883698,0.6509259194135666],"occluded":false},{"timestamp":11.26125,"box":[0.3932291865348816,0.43518516421318054,0.5140625238418579,0.6712962687015533],"occluded":false},{"timestamp":11.5115,"box":[0.4723958671092987,0.4444444477558136,0.5177083685994148,0.6805555522441864],"occluded":false},{"timestamp":11.76175,"box":[0.4723958671092987,0.4018518328666687,0.516145870089531,0.6740740537643433],"occluded":false},{"timestamp":12.012,"box":[0.4755208492279053,0.4435185194015503,0.5223958529531956,0.6546296328306198],"occluded":false},{"timestamp":12.26225,"box":[0.4598958492279053,0.4611110985279083,0.5072916857898235,0.6620370149612427],"occluded":false},{"timestamp":12.5125,"box":[0.48125001788139343,0.5314814448356628,0.49843751825392246,0.6694444119930267],"occluded":false},{"timestamp":12.762749999999999,"box":[0.3932291865348816,0.24722221493721008,0.5223958641290665,0.6370370090007782],"occluded":false},{"timestamp":13.013,"box":[0.47187501192092896,0.4138888716697693,0.494791679084301,0.6444444209337234],"occluded":false},{"timestamp":13.26325,"box":[0.48854169249534607,0.44907405972480774,0.5057291928678751,0.6546296179294586],"occluded":false},{"timestamp":13.513499999999999,"box":[0.483854204416275,0.44907405972480774,0.5015625394880772,0.6148148030042648],"occluded":false},{"timestamp":13.76375,"box":[0.47968751192092896,0.42314815521240234,0.5046875141561031,0.6574074029922485],"occluded":false},{"timestamp":14.014,"box":[0.47343751788139343,0.42222222685813904,0.4979166854172945,0.631481483578682],"occluded":false}],"mean_score":0.8048780560493469},{"noun_phrase":"cyclist in pink jersey","object_id":0,"presentation_id":2,"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}],"mean_score":0.7419355511665344},{"noun_phrase":"cyclist with white helmet","object_id":2,"presentation_id":3,"observations":[{"timestamp":1.5015,"box":[0.444791704416275,0.2370370328426361,0.5750000476837158,0.8046295940876007],"occluded":false},{"timestamp":1.75175,"box":[0.42604169249534607,0.23888888955116272,0.555208370089531,0.817592591047287],"occluded":false},{"timestamp":2.002,"box":[0.4244791865348816,0.24074073135852814,0.5614583641290665,0.7861110717058182],"occluded":false},{"timestamp":2.25225,"box":[0.42500001192092896,0.2490740716457367,0.5635416805744171,0.7750000059604645],"occluded":false},{"timestamp":2.5025,"box":[0.43125003576278687,0.2537037134170532,0.563020870089531,0.8046296238899231],"occluded":false},{"timestamp":2.75275,"box":[0.444791704416275,0.2888888716697693,0.569791704416275,0.7981480956077576],"occluded":false},{"timestamp":3.003,"box":[0.4489583671092987,0.28703704476356506,0.5723958760499954,0.7583333253860474],"occluded":false},{"timestamp":3.25325,"box":[0.4661458432674408,0.25740739703178406,0.5927083492279053,0.7722221910953522],"occluded":false},{"timestamp":3.5035,"box":[0.47864586114883423,0.2620370388031006,0.6072916984558105,0.7935184836387634],"occluded":false},{"timestamp":3.7537499999999997,"box":[0.4739583432674408,0.2638888955116272,0.6020833551883698,0.7833333015441895],"occluded":false},{"timestamp":4.004,"box":[0.47968751192092896,0.2527777850627899,0.6093750149011612,0.7537036836147308],"occluded":false},{"timestamp":4.25425,"box":[0.4713541865348816,0.2509259283542633,0.6046875268220901,0.7898148000240326],"occluded":false},{"timestamp":4.5045,"box":[0.44218751788139343,0.24444444477558136,0.5812500268220901,0.8138888627290726],"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":[0.4515625238418579,0.2212962955236435,0.6015625298023224,0.8148147910833359],"occluded":false},{"timestamp":5.5055,"box":[0.46927085518836975,0.2231481373310089,0.617708370089531,0.8259259164333344],"occluded":false},{"timestamp":5.75575,"box":[0.49427086114883423,0.21666666865348816,0.6427083760499954,0.7870370447635651],"occluded":false},{"timestamp":6.006,"box":[0.5250000357627869,0.20555555820465088,0.6781250387430191,0.8074073791503906],"occluded":false},{"timestamp":6.25625,"box":[0.546875,0.20370370149612427,0.7067708373069763,0.8222222328186035],"occluded":false},{"timestamp":6.5065,"box":[0.5546875,0.2092592567205429,0.7156250029802322,0.783333346247673],"occluded":false},{"timestamp":6.756749999999999,"box":[0.5463541746139526,0.20277777314186096,0.7161458432674408,0.7833333313465118],"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":[0.4869791865348816,0.19074073433876038,0.645833358168602,0.7842592298984528],"occluded":false},{"timestamp":7.75775,"box":[0.460416704416275,0.32222220301628113,0.6250000447034836,0.8259259164333344],"occluded":false},{"timestamp":8.008,"box":[0.43906253576278687,0.32499998807907104,0.6062500476837158,0.845370352268219],"occluded":false},{"timestamp":8.25825,"box":[0.4130208492279053,0.3574073910713196,0.5854166895151138,0.8240740597248077],"occluded":false},{"timestamp":8.5085,"box":[0.39635419845581055,0.10092592239379883,0.6369792073965073,0.8240740299224854],"occluded":false},{"timestamp":8.75875,"box":[0.3515625298023224,0.18611110746860504,0.538541704416275,0.8527777343988419],"occluded":false},{"timestamp":9.009,"box":[0.34166669845581055,0.192592591047287,0.5255208760499954,0.8398148119449615],"occluded":false},{"timestamp":9.25925,"box":[0.3296875059604645,0.1787036955356598,0.5166666805744171,0.807407408952713],"occluded":false},{"timestamp":9.5095,"box":[0.29635417461395264,0.1787036955356598,0.5223958492279053,0.8425925672054291],"occluded":false},{"timestamp":9.75975,"box":[0.30520835518836975,0.12222222238779068,0.49791669845581055,0.8453703299164772],"occluded":false},{"timestamp":10.01,"box":[0.3036458492279053,0.20000000298023224,0.489583358168602,0.8185185343027115],"occluded":false},{"timestamp":10.26025,"box":[0.2796875238418579,0.19722221791744232,0.49375003576278687,0.8185185045003891],"occluded":false},{"timestamp":10.5105,"box":[0.29635417461395264,0.20277777314186096,0.46406251192092896,0.8444444239139557],"occluded":false},{"timestamp":10.76075,"box":[0.31822916865348816,0.2074074000120163,0.49427084624767303,0.8444444388151169],"occluded":false},{"timestamp":11.011,"box":[0.3218750059604645,0.20185184478759766,0.4828125089406967,0.7990740537643433],"occluded":false},{"timestamp":11.26125,"box":[0.3135416805744171,0.19814814627170563,0.48906251788139343,0.8296296149492264],"occluded":false},{"timestamp":11.5115,"box":[0.30781251192092896,0.20555555820465088,0.49427086114883423,0.8490740656852722],"occluded":false},{"timestamp":11.76175,"box":[0.3098958432674408,0.19629628956317902,0.4921875149011612,0.8203703314065933],"occluded":false},{"timestamp":12.012,"box":[0.30677086114883423,0.20555555820465088,0.47708337008953094,0.8185185194015503],"occluded":false},{"timestamp":12.26225,"box":[0.32447919249534607,0.18518517911434174,0.5052083730697632,0.8398148268461227],"occluded":false},{"timestamp":12.5125,"box":[0.3229166865348816,0.18518517911434174,0.49947918951511383,0.7499999850988388],"occluded":false},{"timestamp":12.762749999999999,"box":[0.3229166865348816,0.13611111044883728,0.4921875298023224,0.7907407581806183],"occluded":false},{"timestamp":13.013,"box":[0.31718751788139343,0.1953703612089157,0.47864586114883423,0.8268518298864365],"occluded":false},{"timestamp":13.26325,"box":[0.36666667461395264,0.2074074000120163,0.502083346247673,0.8277777880430222],"occluded":false},{"timestamp":13.513499999999999,"box":[0.33020836114883423,0.21759258210659027,0.48750004172325134,0.8046296089887619],"occluded":false},{"timestamp":13.76375,"box":[0.3307291865348816,0.2203703671693802,0.4812500327825546,0.7888888865709305],"occluded":false},{"timestamp":14.014,"box":[0.33229169249534607,0.22962962090969086,0.476041704416275,0.7999999970197678],"occluded":false}],"mean_score":0.6831682920455933},{"noun_phrase":"cyclist with white helmet","object_id":0,"presentation_id":4,"observations":[{"timestamp":0.0,"box":[0.5869792103767395,0.3722222149372101,0.6781250461935997,0.6888888776302338],"occluded":false},{"timestamp":0.25025,"box":[0.5984375476837158,0.36574074625968933,0.6802083849906921,0.6833333373069763],"occluded":false},{"timestamp":0.5005,"box":[0.5640625357627869,0.37037035822868347,0.6786458715796471,0.6657407283782959],"occluded":false},{"timestamp":0.75075,"box":[0.5963541865348816,0.3759259283542633,0.6750000268220901,0.6574074029922485],"occluded":false},{"timestamp":1.001,"box":[0.5937500596046448,0.3768518567085266,0.668750062584877,0.6657407283782959],"occluded":false},{"timestamp":1.25125,"box":[0.5864583849906921,0.3861111104488373,0.6583333909511566,0.5537036955356598],"occluded":false},{"timestamp":1.5015,"box":[0.5843750238418579,0.39351850748062134,0.65416669100523,0.6472222208976746],"occluded":false},{"timestamp":1.75175,"box":[0.5671875476837158,0.39444443583488464,0.635416716337204,0.6611110866069794],"occluded":false},{"timestamp":2.002,"box":[0.46302086114883423,0.33796295523643494,0.6312500387430191,0.6712962687015533],"occluded":false},{"timestamp":2.25225,"box":[0.47708335518836975,0.35185185074806213,0.6260416954755783,0.6509259343147278],"occluded":false},{"timestamp":2.5025,"box":[0.5593750476837158,0.4037036895751953,0.6218750476837158,0.64907406270504],"occluded":false},{"timestamp":2.75275,"box":[0.5598958730697632,0.4037036895751953,0.6171875409781933,0.6601851582527161],"occluded":false},{"timestamp":3.003,"box":[0.5593750476837158,0.4055555462837219,0.6098958849906921,0.6499999910593033],"occluded":false},{"timestamp":3.25325,"box":[0.5505208373069763,0.412962943315506,0.6072916723787785,0.6398147940635681],"occluded":false},{"timestamp":3.5035,"box":[0.565625011920929,0.4435185194015503,0.603645846247673,0.6129629611968994],"occluded":false},{"timestamp":3.7537499999999997,"box":[0.5640625357627869,0.5435184836387634,0.5750000365078449,0.6527777388691902],"occluded":false},{"timestamp":4.004,"box":[0.5666667222976685,0.4722222089767456,0.5822917222976685,0.587037019431591],"occluded":false},{"timestamp":4.25425,"box":[0.5588542222976685,0.4722222089767456,0.5848958902060986,0.6249999850988388],"occluded":false},{"timestamp":4.5045,"box":[0.5463541746139526,0.49166664481163025,0.5864583440124989,0.6509259045124054],"occluded":false},{"timestamp":4.75475,"box":[0.5390625,0.44722220301628113,0.5864583365619183,0.6518518328666687],"occluded":false},{"timestamp":5.005,"box":[0.5432292222976685,0.4314814805984497,0.5864583924412727,0.635185182094574],"occluded":false},{"timestamp":5.25525,"box":[0.5479167103767395,0.4277777671813965,0.5703125447034836,0.5703703612089157],"occluded":false},{"timestamp":5.5055,"box":[0.5567708611488342,0.42499998211860657,0.5786458626389503,0.5935184955596924],"occluded":false},{"timestamp":5.75575,"box":[0.5770833492279053,0.45092591643333435,0.5968750175088644,0.5296296179294586],"occluded":false},{"timestamp":6.25625,"box":[0.5901042222976685,0.5046296119689941,0.6031250562518835,0.5407407209277153],"occluded":false},{"timestamp":7.25725,"box":[0.6015625596046448,0.43888887763023376,0.6260417271405458,0.527777761220932],"occluded":false},{"timestamp":7.507499999999999,"box":[0.5328125357627869,0.4166666567325592,0.6072917059063911,0.5268518403172493],"occluded":false},{"timestamp":7.75775,"box":[0.5671875476837158,0.4175925850868225,0.59218754991889,0.6444444358348846],"occluded":false},{"timestamp":8.008,"box":[0.5552083849906921,0.4185185134410858,0.5750000532716513,0.6231481432914734],"occluded":false},{"timestamp":8.25825,"box":[0.542187511920929,0.4203703701496124,0.5921875163912773,0.6537037044763565],"occluded":false},{"timestamp":8.75875,"box":[0.5244792103767395,0.4287036955356598,0.5718750469386578,0.6351851671934128],"occluded":false},{"timestamp":9.009,"box":[0.5057291984558105,0.4259259104728699,0.5552083663642406,0.6518518328666687],"occluded":false},{"timestamp":9.25925,"box":[0.5062500238418579,0.4259259104728699,0.5572916939854622,0.6537036895751953],"occluded":false},{"timestamp":9.5095,"box":[0.4911458492279053,0.4296296238899231,0.5536458492279053,0.635185182094574],"occluded":false},{"timestamp":9.75975,"box":[0.5,0.4296296238899231,0.5484375022351742,0.6398148089647293],"occluded":false},{"timestamp":10.01,"box":[0.5005208849906921,0.4296296238899231,0.5489583872258663,0.64907406270504],"occluded":false},{"timestamp":10.26025,"box":[0.483854204416275,0.43611109256744385,0.5317708738148212,0.6416666507720947],"occluded":false},{"timestamp":10.5105,"box":[0.4843750298023224,0.43888887763023376,0.5307292006909847,0.6342592388391495],"occluded":false},{"timestamp":10.76075,"box":[0.5,0.43611109256744385,0.5401041693985462,0.6407407224178314],"occluded":false},{"timestamp":11.011,"box":[0.5036458373069763,0.43981480598449707,0.5401041731238365,0.6370370239019394],"occluded":false},{"timestamp":11.26125,"box":[0.5041667222976685,0.43888887763023376,0.5395833924412727,0.6212962865829468],"occluded":false},{"timestamp":11.5115,"box":[0.5140625238418579,0.4435185194015503,0.5375000257045031,0.635185182094574],"occluded":false},{"timestamp":11.76175,"box":[0.5083333849906921,0.4453703761100769,0.5401042215526104,0.6296296268701553],"occluded":false},{"timestamp":12.012,"box":[0.5135416984558105,0.44907405972480774,0.5406250339001417,0.6222222000360489],"occluded":false},{"timestamp":12.26225,"box":[0.5026041865348816,0.44999998807907104,0.5364583544433117,0.6222222000360489],"occluded":false},{"timestamp":12.5125,"box":[0.499479204416275,0.44814813137054443,0.5328125394880772,0.6277777552604675],"occluded":false},{"timestamp":12.762749999999999,"box":[0.4911458492279053,0.4287036955356598,0.5234375186264515,0.6185185015201569],"occluded":false},{"timestamp":13.013,"box":[0.4911458492279053,0.45185184478759766,0.5213541835546494,0.6166666597127914],"occluded":false},{"timestamp":13.26325,"box":[0.49687501788139343,0.45370370149612427,0.5296875201165676,0.6240740716457367],"occluded":false},{"timestamp":13.513499999999999,"box":[0.4989583492279053,0.4546296298503876,0.5260416846722364,0.6175925880670547],"occluded":false},{"timestamp":13.76375,"box":[0.49635419249534607,0.46388888359069824,0.5239583607763052,0.6064814776182175],"occluded":false},{"timestamp":14.014,"box":[0.4906250238418579,0.460185170173645,0.515104191377759,0.6157407164573669],"occluded":false}],"mean_score":0.8048780560493469}],"dropped":{},"fates":[{"noun_phrase":"cyclist in pink jersey","object_id":0,"frames":11,"mean_score":0.7419355511665344,"mean_area":0.024469083550309804,"outcome":"kept","presentation_id":2,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"cyclist with white helmet","object_id":0,"frames":52,"mean_score":0.8048780560493469,"mean_area":0.01111562797859682,"outcome":"kept","presentation_id":4,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"cyclist with white helmet","object_id":1,"frames":54,"mean_score":0.8048780560493469,"mean_area":0.015123251849840466,"outcome":"kept","presentation_id":1,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"cyclist with white helmet","object_id":2,"frames":51,"mean_score":0.6831682920455933,"mean_area":0.09627511798819198,"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":4,"as_video":true,"fps":1.0,"max_frames":30},"outputs":{"media":"/home/zzk/tmp/rmot_smoke_real_86260/renders/--Dx9HVcQWs_chunk_001_scene_034_choices.mp4","media_bytes":4822999,"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","temperature":0.0,"as_video":true,"media_bytes":4822999,"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 cyclist who drifts furthest back on the right side of the road\n2. the cyclist with a white helmet who rides behind the leader on the left\n3. the cyclist wearing a green helmet","questions":["the cyclist who drifts furthest back on the right side of the road","the cyclist with a white helmet who rides behind the leader on the left","the cyclist wearing a green helmet"]},"outputs":{"raw_response":"{\n  \"answers\": [\n    {\n      \"description_index\": 1,\n      \"matching_ids\": [\n        4\n      ],\n      \"start_frame\": 1,\n      \"end_frame\": 15,\n      \"confidence\": 0.9\n    },\n    {\n      \"description_index\": 2,\n      \"matching_ids\": [\n        1\n      ],\n      \"start_frame\": 1,\n      \"end_frame\": 15,\n      \"confidence\": 1.0\n    },\n    {\n      \"description_index\": 3,\n      \"matching_ids\": [\n        2,\n        3\n      ],\n      \"start_frame\": 1,\n      \"end_frame\": 15,\n      \"confidence\": 0.9\n    }\n  ]\n}","raw_answers":[{"description_index":1,"matching_ids":[4],"start_frame":1,"end_frame":15,"confidence":0.9},{"description_index":2,"matching_ids":[1],"start_frame":1,"end_frame":15,"confidence":1.0},{"description_index":3,"matching_ids":[2,3],"start_frame":1,"end_frame":15,"confidence":0.9}],"answers":[{"expression":"the cyclist who drifts furthest back on the right side of the road","matching_presentation_ids":[4],"start_choice_frame":1,"end_choice_frame":15,"confidence":0.9,"self_reported_confidence":0.9},{"expression":"the cyclist with a white helmet who rides behind the leader on the left","matching_presentation_ids":[1],"start_choice_frame":1,"end_choice_frame":15,"confidence":1.0,"self_reported_confidence":1.0},{"expression":"the cyclist wearing a green helmet","matching_presentation_ids":[2,3],"start_choice_frame":1,"end_choice_frame":15,"confidence":0.9,"self_reported_confidence":0.9}]}},"verify":{"inputs":{"expressions":3,"tracklets":4,"selections":3,"gates":"interval, count, unknown ids, confidence, then geometry (motion)"},"outputs":{"kept":[{"expression":"the cyclist who drifts furthest back on the right side of the road","kind":"motion","is_negative":false,"tracks":[{"timestamps":[0.0,0.25025,0.5005,0.75075,1.001,1.25125,1.5015,1.75175,2.002,2.25225,2.5025,2.75275,3.003,3.25325,3.5035,3.7537499999999997,4.004,4.25425,4.5045,4.75475,5.005,5.25525,5.5055,5.75575,6.25625,7.25725,7.507499999999999,7.75775,8.008,8.25825,8.75875,9.009,9.25925,9.5095,9.75975,10.01,10.26025,10.5105,10.76075,11.011,11.26125,11.5115,11.76175,12.012,12.26225,12.5125,12.762749999999999,13.013,13.26325,13.513499999999999,13.76375,14.014],"boxes":[[0.5869792103767395,0.3722222149372101,0.6781250461935997,0.6888888776302338],[0.5984375476837158,0.36574074625968933,0.6802083849906921,0.6833333373069763],[0.5640625357627869,0.37037035822868347,0.6786458715796471,0.6657407283782959],[0.5963541865348816,0.3759259283542633,0.6750000268220901,0.6574074029922485],[0.5937500596046448,0.3768518567085266,0.668750062584877,0.6657407283782959],[0.5864583849906921,0.3861111104488373,0.6583333909511566,0.5537036955356598],[0.5843750238418579,0.39351850748062134,0.65416669100523,0.6472222208976746],[0.5671875476837158,0.39444443583488464,0.635416716337204,0.6611110866069794],[0.46302086114883423,0.33796295523643494,0.6312500387430191,0.6712962687015533],[0.47708335518836975,0.35185185074806213,0.6260416954755783,0.6509259343147278],[0.5593750476837158,0.4037036895751953,0.6218750476837158,0.64907406270504],[0.5598958730697632,0.4037036895751953,0.6171875409781933,0.6601851582527161],[0.5593750476837158,0.4055555462837219,0.6098958849906921,0.6499999910593033],[0.5505208373069763,0.412962943315506,0.6072916723787785,0.6398147940635681],[0.565625011920929,0.4435185194015503,0.603645846247673,0.6129629611968994],[0.5640625357627869,0.5435184836387634,0.5750000365078449,0.6527777388691902],[0.5666667222976685,0.4722222089767456,0.5822917222976685,0.587037019431591],[0.5588542222976685,0.4722222089767456,0.5848958902060986,0.6249999850988388],[0.5463541746139526,0.49166664481163025,0.5864583440124989,0.6509259045124054],[0.5390625,0.44722220301628113,0.5864583365619183,0.6518518328666687],[0.5432292222976685,0.4314814805984497,0.5864583924412727,0.635185182094574],[0.5479167103767395,0.4277777671813965,0.5703125447034836,0.5703703612089157],[0.5567708611488342,0.42499998211860657,0.5786458626389503,0.5935184955596924],[0.5770833492279053,0.45092591643333435,0.5968750175088644,0.5296296179294586],[0.5901042222976685,0.5046296119689941,0.6031250562518835,0.5407407209277153],[0.6015625596046448,0.43888887763023376,0.6260417271405458,0.527777761220932],[0.5328125357627869,0.4166666567325592,0.6072917059063911,0.5268518403172493],[0.5671875476837158,0.4175925850868225,0.59218754991889,0.6444444358348846],[0.5552083849906921,0.4185185134410858,0.5750000532716513,0.6231481432914734],[0.542187511920929,0.4203703701496124,0.5921875163912773,0.6537037044763565],[0.5244792103767395,0.4287036955356598,0.5718750469386578,0.6351851671934128],[0.5057291984558105,0.4259259104728699,0.5552083663642406,0.6518518328666687],[0.5062500238418579,0.4259259104728699,0.5572916939854622,0.6537036895751953],[0.4911458492279053,0.4296296238899231,0.5536458492279053,0.635185182094574],[0.5,0.4296296238899231,0.5484375022351742,0.6398148089647293],[0.5005208849906921,0.4296296238899231,0.5489583872258663,0.64907406270504],[0.483854204416275,0.43611109256744385,0.5317708738148212,0.6416666507720947],[0.4843750298023224,0.43888887763023376,0.5307292006909847,0.6342592388391495],[0.5,0.43611109256744385,0.5401041693985462,0.6407407224178314],[0.5036458373069763,0.43981480598449707,0.5401041731238365,0.6370370239019394],[0.5041667222976685,0.43888887763023376,0.5395833924412727,0.6212962865829468],[0.5140625238418579,0.4435185194015503,0.5375000257045031,0.635185182094574],[0.5083333849906921,0.4453703761100769,0.5401042215526104,0.6296296268701553],[0.5135416984558105,0.44907405972480774,0.5406250339001417,0.6222222000360489],[0.5026041865348816,0.44999998807907104,0.5364583544433117,0.6222222000360489],[0.499479204416275,0.44814813137054443,0.5328125394880772,0.6277777552604675],[0.4911458492279053,0.4287036955356598,0.5234375186264515,0.6185185015201569],[0.4911458492279053,0.45185184478759766,0.5213541835546494,0.6166666597127914],[0.49687501788139343,0.45370370149612427,0.5296875201165676,0.6240740716457367],[0.4989583492279053,0.4546296298503876,0.5260416846722364,0.6175925880670547],[0.49635419249534607,0.46388888359069824,0.5239583607763052,0.6064814776182175],[0.4906250238418579,0.460185170173645,0.515104191377759,0.6157407164573669]]}]},{"expression":"the cyclist with a white helmet who rides behind the leader on the left","kind":"spatial","is_negative":false,"tracks":[{"timestamps":[0.0,0.25025,0.5005,0.75075,1.001,1.25125,1.5015,1.75175,2.002,2.25225,2.5025,2.75275,3.003,3.25325,3.5035,3.7537499999999997,4.004,4.25425,4.5045,4.75475,5.005,5.25525,5.5055,5.75575,6.006,6.25625,6.5065,6.756749999999999,7.007,7.25725,7.507499999999999,7.75775,8.008,8.25825,8.5085,9.009,9.75975,10.01,10.26025,10.5105,10.76075,11.011,11.26125,11.5115,11.76175,12.012,12.26225,12.5125,12.762749999999999,13.013,13.26325,13.513499999999999,13.76375,14.014],"boxes":[[0.4385416805744171,0.412962943315506,0.4802083484828472,0.6861110925674438],[0.4385416805744171,0.4046296179294586,0.4911458492279053,0.6833333075046539],[0.42760419845581055,0.4000000059604645,0.48750003427267075,0.6842592656612396],[0.41822919249534607,0.4055555462837219,0.4776041954755783,0.6712962687015533],[0.41041669249534607,0.4037036895751953,0.4635416939854622,0.6888888776302338],[0.40260419249534607,0.4037036895751953,0.4682291969656944,0.6925925612449646],[0.40000003576278687,0.4009259045124054,0.4697917029261589,0.6611110866069794],[0.39010417461395264,0.4037036895751953,0.45000001043081284,0.6601851582527161],[0.39010417461395264,0.40740740299224854,0.45000001043081284,0.6527777761220932],[0.3958333432674408,0.41111111640930176,0.4567708447575569,0.6842592656612396],[0.40416669845581055,0.4175925850868225,0.4567708671092987,0.6574074029922485],[0.4098958671092987,0.4175925850868225,0.460416704416275,0.6574074029922485],[0.40937501192092896,0.42222222685813904,0.4651041813194752,0.6740740835666656],[0.4203125238418579,0.42407405376434326,0.4713541939854622,0.6685184985399246],[0.42760419845581055,0.42407405376434326,0.48072919994592667,0.6527777463197708],[0.42500001192092896,0.4268518388271332,0.6020833551883698,0.6685184985399246],[0.43802085518836975,0.42499998211860657,0.5744791924953461,0.6675925701856613],[0.44218751788139343,0.4268518388271332,0.49010418727993965,0.6509259045124054],[0.4375000298023224,0.4287036955356598,0.4614583645015955,0.6259259134531021],[0.444791704416275,0.4416666626930237,0.47447920590639114,0.6703703552484512],[0.45104169845581055,0.460185170173645,0.47708336636424065,0.6398147940635681],[0.4515625238418579,0.4166666567325592,0.479687524959445,0.6416666507720947],[0.45468753576278687,0.41018518805503845,0.482291704043746,0.6805555522441864],[0.45625001192092896,0.40648147463798523,0.5078125149011612,0.6787036955356598],[0.46302086114883423,0.39814814925193787,0.602083370089531,0.6638888716697693],[0.46666669845581055,0.39444443583488464,0.5302083715796471,0.6749999821186066],[0.47187501192092896,0.39537036418914795,0.6161458492279053,0.6777777671813965],[0.47343751788139343,0.39351850748062134,0.6026041954755783,0.6851851642131805],[0.46302086114883423,0.39259257912635803,0.6437500417232513,0.6833333075046539],[0.4619791805744171,0.3916666507720947,0.5187500156462193,0.6638888716697693],[0.4520833492279053,0.3907407224178314,0.5333333536982536,0.6898148059844971],[0.44062501192092896,0.44814813137054443,0.4723958484828472,0.6361110955476761],[0.43385419249534607,0.46759259700775146,0.46406252682209015,0.6833333373069763],[0.4479166865348816,0.550000011920929,0.46197918709367514,0.6157407537102699],[0.40052086114883423,0.54537034034729,0.43177086114883423,0.6249999701976776],[0.3932291865348816,0.47592592239379883,0.4088541865348816,0.5722222179174423],[0.3812500238418579,0.46203702688217163,0.4791666939854622,0.7259259223937988],[0.46302086114883423,0.45092591643333435,0.4875000286847353,0.6703703552484512],[0.4453125298023224,0.4453703761100769,0.47968753054738045,0.6388888955116272],[0.36927086114883423,0.4546296298503876,0.48489587008953094,0.6888888776302338],[0.4755208492279053,0.44999998807907104,0.5020833499729633,0.6416666507720947],[0.4791666865348816,0.45092591643333435,0.5083333551883698,0.6509259194135666],[0.3932291865348816,0.43518516421318054,0.5140625238418579,0.6712962687015533],[0.4723958671092987,0.4444444477558136,0.5177083685994148,0.6805555522441864],[0.4723958671092987,0.4018518328666687,0.516145870089531,0.6740740537643433],[0.4755208492279053,0.4435185194015503,0.5223958529531956,0.6546296328306198],[0.4598958492279053,0.4611110985279083,0.5072916857898235,0.6620370149612427],[0.48125001788139343,0.5314814448356628,0.49843751825392246,0.6694444119930267],[0.3932291865348816,0.24722221493721008,0.5223958641290665,0.6370370090007782],[0.47187501192092896,0.4138888716697693,0.494791679084301,0.6444444209337234],[0.48854169249534607,0.44907405972480774,0.5057291928678751,0.6546296179294586],[0.483854204416275,0.44907405972480774,0.5015625394880772,0.6148148030042648],[0.47968751192092896,0.42314815521240234,0.5046875141561031,0.6574074029922485],[0.47343751788139343,0.42222222685813904,0.4979166854172945,0.631481483578682]]}]}],"rejected":{"the cyclist wearing a green helmet":"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/--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 cyclist who drifts furthest back on the right side of the road","kind":"motion","is_negative":false,"tracks":[{"timestamps":[0.0,0.25025,0.5005,0.75075,1.001,1.25125,1.5015,1.75175,2.002,2.25225,2.5025,2.75275,3.003,3.25325,3.5035,3.7537499999999997,4.004,4.25425,4.5045,4.75475,5.005,5.25525,5.5055,5.75575,6.25625,7.25725,7.507499999999999,7.75775,8.008,8.25825,8.75875,9.009,9.25925,9.5095,9.75975,10.01,10.26025,10.5105,10.76075,11.011,11.26125,11.5115,11.76175,12.012,12.26225,12.5125,12.762749999999999,13.013,13.26325,13.513499999999999,13.76375,14.014],"boxes":[[0.5869792103767395,0.3722222149372101,0.6781250461935997,0.6888888776302338],[0.5984375476837158,0.36574074625968933,0.6802083849906921,0.6833333373069763],[0.5640625357627869,0.37037035822868347,0.6786458715796471,0.6657407283782959],[0.5963541865348816,0.3759259283542633,0.6750000268220901,0.6574074029922485],[0.5937500596046448,0.3768518567085266,0.668750062584877,0.6657407283782959],[0.5864583849906921,0.3861111104488373,0.6583333909511566,0.5537036955356598],[0.5843750238418579,0.39351850748062134,0.65416669100523,0.6472222208976746],[0.5671875476837158,0.39444443583488464,0.635416716337204,0.6611110866069794],[0.46302086114883423,0.33796295523643494,0.6312500387430191,0.6712962687015533],[0.47708335518836975,0.35185185074806213,0.6260416954755783,0.6509259343147278],[0.5593750476837158,0.4037036895751953,0.6218750476837158,0.64907406270504],[0.5598958730697632,0.4037036895751953,0.6171875409781933,0.6601851582527161],[0.5593750476837158,0.4055555462837219,0.6098958849906921,0.6499999910593033],[0.5505208373069763,0.412962943315506,0.6072916723787785,0.6398147940635681],[0.565625011920929,0.4435185194015503,0.603645846247673,0.6129629611968994],[0.5640625357627869,0.5435184836387634,0.5750000365078449,0.6527777388691902],[0.5666667222976685,0.4722222089767456,0.5822917222976685,0.587037019431591],[0.5588542222976685,0.4722222089767456,0.5848958902060986,0.6249999850988388],[0.5463541746139526,0.49166664481163025,0.5864583440124989,0.6509259045124054],[0.5390625,0.44722220301628113,0.5864583365619183,0.6518518328666687],[0.5432292222976685,0.4314814805984497,0.5864583924412727,0.635185182094574],[0.5479167103767395,0.4277777671813965,0.5703125447034836,0.5703703612089157],[0.5567708611488342,0.42499998211860657,0.5786458626389503,0.5935184955596924],[0.5770833492279053,0.45092591643333435,0.5968750175088644,0.5296296179294586],[0.5901042222976685,0.5046296119689941,0.6031250562518835,0.5407407209277153],[0.6015625596046448,0.43888887763023376,0.6260417271405458,0.527777761220932],[0.5328125357627869,0.4166666567325592,0.6072917059063911,0.5268518403172493],[0.5671875476837158,0.4175925850868225,0.59218754991889,0.6444444358348846],[0.5552083849906921,0.4185185134410858,0.5750000532716513,0.6231481432914734],[0.542187511920929,0.4203703701496124,0.5921875163912773,0.6537037044763565],[0.5244792103767395,0.4287036955356598,0.5718750469386578,0.6351851671934128],[0.5057291984558105,0.4259259104728699,0.5552083663642406,0.6518518328666687],[0.5062500238418579,0.4259259104728699,0.5572916939854622,0.6537036895751953],[0.4911458492279053,0.4296296238899231,0.5536458492279053,0.635185182094574],[0.5,0.4296296238899231,0.5484375022351742,0.6398148089647293],[0.5005208849906921,0.4296296238899231,0.5489583872258663,0.64907406270504],[0.483854204416275,0.43611109256744385,0.5317708738148212,0.6416666507720947],[0.4843750298023224,0.43888887763023376,0.5307292006909847,0.6342592388391495],[0.5,0.43611109256744385,0.5401041693985462,0.6407407224178314],[0.5036458373069763,0.43981480598449707,0.5401041731238365,0.6370370239019394],[0.5041667222976685,0.43888887763023376,0.5395833924412727,0.6212962865829468],[0.5140625238418579,0.4435185194015503,0.5375000257045031,0.635185182094574],[0.5083333849906921,0.4453703761100769,0.5401042215526104,0.6296296268701553],[0.5135416984558105,0.44907405972480774,0.5406250339001417,0.6222222000360489],[0.5026041865348816,0.44999998807907104,0.5364583544433117,0.6222222000360489],[0.499479204416275,0.44814813137054443,0.5328125394880772,0.6277777552604675],[0.4911458492279053,0.4287036955356598,0.5234375186264515,0.6185185015201569],[0.4911458492279053,0.45185184478759766,0.5213541835546494,0.6166666597127914],[0.49687501788139343,0.45370370149612427,0.5296875201165676,0.6240740716457367],[0.4989583492279053,0.4546296298503876,0.5260416846722364,0.6175925880670547],[0.49635419249534607,0.46388888359069824,0.5239583607763052,0.6064814776182175],[0.4906250238418579,0.460185170173645,0.515104191377759,0.6157407164573669]]}]},{"expression":"the cyclist with a white helmet who rides behind the leader on the left","kind":"spatial","is_negative":false,"tracks":[{"timestamps":[0.0,0.25025,0.5005,0.75075,1.001,1.25125,1.5015,1.75175,2.002,2.25225,2.5025,2.75275,3.003,3.25325,3.5035,3.7537499999999997,4.004,4.25425,4.5045,4.75475,5.005,5.25525,5.5055,5.75575,6.006,6.25625,6.5065,6.756749999999999,7.007,7.25725,7.507499999999999,7.75775,8.008,8.25825,8.5085,9.009,9.75975,10.01,10.26025,10.5105,10.76075,11.011,11.26125,11.5115,11.76175,12.012,12.26225,12.5125,12.762749999999999,13.013,13.26325,13.513499999999999,13.76375,14.014],"boxes":[[0.4385416805744171,0.412962943315506,0.4802083484828472,0.6861110925674438],[0.4385416805744171,0.4046296179294586,0.4911458492279053,0.6833333075046539],[0.42760419845581055,0.4000000059604645,0.48750003427267075,0.6842592656612396],[0.41822919249534607,0.4055555462837219,0.4776041954755783,0.6712962687015533],[0.41041669249534607,0.4037036895751953,0.4635416939854622,0.6888888776302338],[0.40260419249534607,0.4037036895751953,0.4682291969656944,0.6925925612449646],[0.40000003576278687,0.4009259045124054,0.4697917029261589,0.6611110866069794],[0.39010417461395264,0.4037036895751953,0.45000001043081284,0.6601851582527161],[0.39010417461395264,0.40740740299224854,0.45000001043081284,0.6527777761220932],[0.3958333432674408,0.41111111640930176,0.4567708447575569,0.6842592656612396],[0.40416669845581055,0.4175925850868225,0.4567708671092987,0.6574074029922485],[0.4098958671092987,0.4175925850868225,0.460416704416275,0.6574074029922485],[0.40937501192092896,0.42222222685813904,0.4651041813194752,0.6740740835666656],[0.4203125238418579,0.42407405376434326,0.4713541939854622,0.6685184985399246],[0.42760419845581055,0.42407405376434326,0.48072919994592667,0.6527777463197708],[0.42500001192092896,0.4268518388271332,0.6020833551883698,0.6685184985399246],[0.43802085518836975,0.42499998211860657,0.5744791924953461,0.6675925701856613],[0.44218751788139343,0.4268518388271332,0.49010418727993965,0.6509259045124054],[0.4375000298023224,0.4287036955356598,0.4614583645015955,0.6259259134531021],[0.444791704416275,0.4416666626930237,0.47447920590639114,0.6703703552484512],[0.45104169845581055,0.460185170173645,0.47708336636424065,0.6398147940635681],[0.4515625238418579,0.4166666567325592,0.479687524959445,0.6416666507720947],[0.45468753576278687,0.41018518805503845,0.482291704043746,0.6805555522441864],[0.45625001192092896,0.40648147463798523,0.5078125149011612,0.6787036955356598],[0.46302086114883423,0.39814814925193787,0.602083370089531,0.6638888716697693],[0.46666669845581055,0.39444443583488464,0.5302083715796471,0.6749999821186066],[0.47187501192092896,0.39537036418914795,0.6161458492279053,0.6777777671813965],[0.47343751788139343,0.39351850748062134,0.6026041954755783,0.6851851642131805],[0.46302086114883423,0.39259257912635803,0.6437500417232513,0.6833333075046539],[0.4619791805744171,0.3916666507720947,0.5187500156462193,0.6638888716697693],[0.4520833492279053,0.3907407224178314,0.5333333536982536,0.6898148059844971],[0.44062501192092896,0.44814813137054443,0.4723958484828472,0.6361110955476761],[0.43385419249534607,0.46759259700775146,0.46406252682209015,0.6833333373069763],[0.4479166865348816,0.550000011920929,0.46197918709367514,0.6157407537102699],[0.40052086114883423,0.54537034034729,0.43177086114883423,0.6249999701976776],[0.3932291865348816,0.47592592239379883,0.4088541865348816,0.5722222179174423],[0.3812500238418579,0.46203702688217163,0.4791666939854622,0.7259259223937988],[0.46302086114883423,0.45092591643333435,0.4875000286847353,0.6703703552484512],[0.4453125298023224,0.4453703761100769,0.47968753054738045,0.6388888955116272],[0.36927086114883423,0.4546296298503876,0.48489587008953094,0.6888888776302338],[0.4755208492279053,0.44999998807907104,0.5020833499729633,0.6416666507720947],[0.4791666865348816,0.45092591643333435,0.5083333551883698,0.6509259194135666],[0.3932291865348816,0.43518516421318054,0.5140625238418579,0.6712962687015533],[0.4723958671092987,0.4444444477558136,0.5177083685994148,0.6805555522441864],[0.4723958671092987,0.4018518328666687,0.516145870089531,0.6740740537643433],[0.4755208492279053,0.4435185194015503,0.5223958529531956,0.6546296328306198],[0.4598958492279053,0.4611110985279083,0.5072916857898235,0.6620370149612427],[0.48125001788139343,0.5314814448356628,0.49843751825392246,0.6694444119930267],[0.3932291865348816,0.24722221493721008,0.5223958641290665,0.6370370090007782],[0.47187501192092896,0.4138888716697693,0.494791679084301,0.6444444209337234],[0.48854169249534607,0.44907405972480774,0.5057291928678751,0.6546296179294586],[0.483854204416275,0.44907405972480774,0.5015625394880772,0.6148148030042648],[0.47968751192092896,0.42314815521240234,0.5046875141561031,0.6574074029922485],[0.47343751788139343,0.42222222685813904,0.4979166854172945,0.631481483578682]]}]}],"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","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 dreadlocks and a beard\",\n      \"noun_phrase\": \"man with beard\",\n      \"kind\": \"appearance\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the woman who raises her hand to shield her eyes from the sun\",\n      \"noun_phrase\": \"woman\",\n      \"kind\": \"motion\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the man wearing a blue cap\",\n      \"noun_phrase\": \"man\",\n      \"kind\": \"appearance\",\n      \"expected_count\": 0,\n      \"is_negative\": true\n    }\n  ]\n}","raw_items":[{"expression":"the man with dreadlocks and a beard","noun_phrase":"man with beard","kind":"appearance","expected_count":1,"is_negative":false},{"expression":"the woman who raises her hand to shield her eyes from the sun","noun_phrase":"woman","kind":"motion","expected_count":1,"is_negative":false},{"expression":"the man wearing a blue cap","noun_phrase":"man","kind":"appearance","expected_count":0,"is_negative":true}],"accepted":[{"expression":"the man with dreadlocks and a beard","noun_phrase":"man with beard","kind":"appearance","expected_count":1,"is_negative":false},{"expression":"the woman who raises her hand to shield her eyes from the sun","noun_phrase":"woman","kind":"motion","expected_count":1,"is_negative":false},{"expression":"the man wearing a blue cap","noun_phrase":"man","kind":"appearance","expected_count":0,"is_negative":true}],"gated":[]}},"track":{"inputs":{"tracker":"Sam31MultiplexTracker","model_id":"/mnt/ceph3/ec/jacky/sam3.1","noun_phrases":["man","man with 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":"man","object_id":0,"observations":[{"timestamp":0.0,"box":[0.5562500357627869,0.07962962985038757,0.9994792342185974,0.9703703820705414],"occluded":false},{"timestamp":0.26,"box":[0.34322917461395264,0.20462962985038757,0.9708333611488342,0.9722221791744232],"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":[0.4515625238418579,0.10648147761821747,0.9885417222976685,0.9722221940755844],"occluded":false},{"timestamp":1.3,"box":[0.4401041865348816,0.1101851835846901,0.9708333611488342,0.9722221866250038],"occluded":false},{"timestamp":1.56,"box":[0.4609375298023224,0.043518517166376114,0.999479204416275,0.9722221828997135],"occluded":false},{"timestamp":1.82,"box":[0.3661458492279053,0.13518518209457397,0.9895833730697632,0.9722222089767456],"occluded":false},{"timestamp":2.08,"box":[0.45625001192092896,0.1944444477558136,0.9781250357627869,0.972222238779068],"occluded":false},{"timestamp":2.34,"box":[0.39375001192092896,0.10740740597248077,0.9755208492279053,0.9722222238779068],"occluded":false},{"timestamp":2.6,"box":[0.43541669845581055,0.09259258955717087,0.970833420753479,0.972222201526165],"occluded":false},{"timestamp":2.86,"box":[0.4437500238418579,0.13518518209457397,0.9729167222976685,0.9722222089767456],"occluded":false},{"timestamp":3.12,"box":[0.44166669249534607,0.14907407760620117,0.9828125536441803,0.9722222089767456],"occluded":false},{"timestamp":3.38,"box":[0.45625001192092896,0.12222222238779068,0.9994792342185974,0.9722221866250038],"occluded":false},{"timestamp":3.64,"box":[0.45781251788139343,0.1268518567085266,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":3.9,"box":[0.45885419845581055,0.1268518567085266,0.9994792342185974,0.9722222089767456],"occluded":false},{"timestamp":4.16,"box":[0.4593750238418579,0.15833333134651184,0.9994792342185974,0.9722221791744232],"occluded":false},{"timestamp":4.42,"box":[0.45781251788139343,0.10925925523042679,0.999479204416275,0.9722222164273262],"occluded":false},{"timestamp":4.68,"box":[0.29635417461395264,0.08055555075407028,0.9994792342185974,0.972222201526165],"occluded":false},{"timestamp":4.94,"box":[0.4364583492279053,0.12962962687015533,0.9994792342185974,0.9722222238779068],"occluded":false},{"timestamp":5.2,"box":[0.31041666865348816,0.09351851791143417,0.999479204416275,0.9731481298804283],"occluded":false},{"timestamp":5.46,"box":[0.2828125059604645,0.07037036865949631,0.999479204416275,0.972222201526165],"occluded":false},{"timestamp":5.72,"box":[0.4505208432674408,0.10740740597248077,0.999479204416275,0.9722222238779068],"occluded":false},{"timestamp":5.98,"box":[0.3661458492279053,0.12962962687015533,0.9937500357627869,0.9722222238779068],"occluded":false},{"timestamp":6.24,"box":[0.24114584922790527,0.13333332538604736,0.9994792342185974,0.9990740418434143],"occluded":false},{"timestamp":6.5,"box":[0.33802086114883423,0.14722222089767456,0.9994792342185974,0.9731481075286865],"occluded":false},{"timestamp":6.76,"box":[0.33802086114883423,0.12037036567926407,0.9994792342185974,0.9722221866250038],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.3656250238418579,0.14444443583488464,0.9994792342185974,0.9722221791744232],"occluded":false},{"timestamp":7.28,"box":[0.31041666865348816,0.13425925374031067,0.999479204416275,0.9722221791744232],"occluded":false},{"timestamp":7.54,"box":[0.3098958432674408,0.0962962955236435,0.999479204416275,0.9722221940755844],"occluded":false},{"timestamp":7.8,"box":[0.29635417461395264,0.09351851791143417,0.9994792342185974,0.9722222313284874],"occluded":false},{"timestamp":8.06,"box":[0.24114584922790527,0.08796296268701553,0.9994792342185974,0.9722221866250038],"occluded":false},{"timestamp":8.32,"box":[0.4661458432674408,0.13703703880310059,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":8.58,"box":[0.46145835518836975,0.09166666865348816,0.999479204416275,0.9722221791744232],"occluded":false},{"timestamp":8.84,"box":[0.4593750238418579,0.09074074029922485,0.9994792342185974,0.9722222089767456],"occluded":false},{"timestamp":9.1,"box":[0.2828125059604645,0.07685185223817825,0.999479204416275,0.9722222164273262],"occluded":false},{"timestamp":9.36,"box":[0.4505208432674408,0.060185182839632034,0.999479204416275,0.9731481559574604],"occluded":false},{"timestamp":9.620000000000001,"box":[0.37552085518836975,0.14907407760620117,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":9.88,"box":[0.37187501788139343,0.1546296328306198,0.999479204416275,0.9722221940755844],"occluded":false},{"timestamp":10.14,"box":[0.11614584177732468,0.13703703880310059,0.9968750402331352,0.9722222089767456],"occluded":false},{"timestamp":10.4,"box":[0.03281250223517418,0.13703703880310059,0.9994792006909847,0.9722222089767456],"occluded":false},{"timestamp":10.66,"box":[0.3661458492279053,0.19907407462596893,0.9994792342185974,0.9722221940755844],"occluded":false},{"timestamp":10.92,"box":[0.32447919249534607,0.14537036418914795,0.999479204416275,0.9731481075286865],"occluded":false},{"timestamp":11.18,"box":[0.4359375238418579,0.13518518209457397,0.9994792342185974,0.9722222089767456],"occluded":false},{"timestamp":11.44,"box":[0.4520833492279053,0.14074073731899261,0.9994792342185974,0.9722221940755844],"occluded":false},{"timestamp":11.700000000000001,"box":[0.44739586114883423,0.14537036418914795,0.9994792342185974,0.9722222089767456],"occluded":false},{"timestamp":11.96,"box":[0.4453125298023224,0.1675925850868225,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":12.22,"box":[0.4635416865348816,0.1546296328306198,0.9994792342185974,0.9722221940755844],"occluded":false},{"timestamp":12.48,"box":[0.4598958492279053,0.13611111044883728,0.9994792342185974,0.9722221791744232],"occluded":false},{"timestamp":12.74,"box":[0.45000001788139343,0.1398148089647293,0.999479204416275,0.9722222238779068],"occluded":false},{"timestamp":13.0,"box":[0.45468753576278687,0.1268518567085266,0.9994792342185974,0.9722222089767456],"occluded":false},{"timestamp":13.26,"box":[0.46562501788139343,0.13055555522441864,0.999479204416275,0.9722221940755844],"occluded":false},{"timestamp":13.52,"box":[0.4505208432674408,0.11203703284263611,0.999479204416275,0.9722221791744232],"occluded":false},{"timestamp":13.780000000000001,"box":[0.3765625059604645,0.07037036865949631,0.999479204416275,0.972222201526165],"occluded":false},{"timestamp":14.040000000000001,"box":[0.35104167461395264,0.06481481343507767,0.9994792342185974,0.9722222164273262],"occluded":false},{"timestamp":14.3,"box":[0.21302084624767303,0.05925925821065903,0.9994792193174362,0.9722222313284874],"occluded":false},{"timestamp":14.56,"box":[0.4817708730697632,0.08148147910833359,0.9994792342185974,0.9722222313284874],"occluded":false},{"timestamp":14.82,"box":[0.4833333492279053,0.10092592239379883,0.9994792342185974,0.9722222089767456],"occluded":false},{"timestamp":15.08,"box":[0.4755208492279053,0.12962962687015533,0.9994792342185974,0.9722222238779068],"occluded":false},{"timestamp":15.34,"box":[0.452604204416275,0.14351852238178253,0.999479204416275,0.9722222238779068],"occluded":false},{"timestamp":15.6,"box":[0.06041666865348816,0.15648147463798523,0.999479204416275,0.9722221791744232],"occluded":false},{"timestamp":15.860000000000001,"box":[0.45468753576278687,0.1509259194135666,0.9994792342185974,0.9722221940755844],"occluded":false},{"timestamp":16.12,"box":[0.13333334028720856,0.16203702986240387,0.9994791895151138,0.9722222238779068],"occluded":false},{"timestamp":16.38,"box":[0.46562501788139343,0.15648147463798523,0.999479204416275,0.9722221791744232],"occluded":false},{"timestamp":16.64,"box":[0.4593750238418579,0.15833333134651184,0.9994792342185974,0.9722221791744232],"occluded":false},{"timestamp":16.9,"box":[0.46406251192092896,0.15925925970077515,0.9994792342185974,0.9712963104248047],"occluded":false},{"timestamp":17.16,"box":[0.46927085518836975,0.16296295821666718,0.999479204416275,0.9722221940755844],"occluded":false},{"timestamp":17.42,"box":[0.4661458432674408,0.1546296328306198,0.999479204416275,0.9722221940755844],"occluded":false},{"timestamp":17.68,"box":[0.47708335518836975,0.14907407760620117,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":17.94,"box":[0.2828125059604645,0.1537037044763565,0.999479204416275,0.9722222238779068],"occluded":false},{"timestamp":18.2,"box":[0.07447917014360428,0.12129629403352737,0.999479241669178,0.9722222164273262],"occluded":false},{"timestamp":18.46,"box":[0.03281250223517418,0.03611110895872116,0.9994792006909847,0.9712963029742241],"occluded":false},{"timestamp":18.72,"box":[0.10208334028720856,0.031481482088565826,0.9994792491197586,0.9722221866250038],"occluded":false},{"timestamp":18.98,"box":[0.4286458492279053,0.03333333134651184,0.9994792342185974,0.9712962806224823],"occluded":false},{"timestamp":19.240000000000002,"box":[0.43177086114883423,0.09259258955717087,0.9994792342185974,0.972222201526165],"occluded":false},{"timestamp":19.5,"box":[0.4359375238418579,0.1398148089647293,0.9994792342185974,0.9722222238779068],"occluded":false},{"timestamp":19.76,"box":[0.43281251192092896,0.18333333730697632,0.9994792342185974,0.9722222089767456],"occluded":false},{"timestamp":20.02,"box":[0.4442708492279053,0.18796296417713165,0.9994792342185974,0.9722222238779068],"occluded":false},{"timestamp":20.28,"box":[0.4567708671092987,0.17037037014961243,0.999479204416275,0.9722221791744232],"occluded":false}],"scores":[0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846,0.949367105960846]},{"noun_phrase":"man","object_id":1,"observations":[{"timestamp":1.3,"box":[0.4046875238418579,0.6564814448356628,0.425000024959445,0.758333295583725],"occluded":false},{"timestamp":1.56,"box":[0.3593750298023224,0.6879629492759705,0.38072919845581055,0.7925925776362419],"occluded":false},{"timestamp":1.82,"box":[0.3541666865348816,0.6814814805984497,0.3744791876524687,0.7870370373129845],"occluded":false},{"timestamp":2.08,"box":[0.37916669249534607,0.690740704536438,0.4000000264495611,0.7953703328967094],"occluded":false},{"timestamp":2.34,"box":[0.41875001788139343,0.6833333373069763,0.4395833518356085,0.7861111164093018],"occluded":false},{"timestamp":2.6,"box":[0.45625001192092896,0.6824073791503906,0.475520845502615,0.7842592298984528],"occluded":false},{"timestamp":3.38,"box":[0.4541666805744171,0.6796296238899231,0.47395834885537624,0.7787036970257759],"occluded":false},{"timestamp":3.64,"box":[0.3895833492279053,0.6916666626930237,0.4114583507180214,0.7944444417953491],"occluded":false},{"timestamp":3.9,"box":[0.3500000238418579,0.7037037014961243,0.37343752570450306,0.8101851791143417],"occluded":false},{"timestamp":4.16,"box":[0.3270833492279053,0.7018518447875977,0.34947918355464935,0.8092592507600784],"occluded":false},{"timestamp":4.42,"box":[0.3109375238418579,0.7092592716217041,0.334895858541131,0.8194444552063942],"occluded":false},{"timestamp":4.68,"box":[0.3062500059604645,0.6953703761100769,0.32864584028720856,0.8046296313405037],"occluded":false},{"timestamp":4.94,"box":[0.32864585518836975,0.6796296238899231,0.3520833570510149,0.7851851806044579],"occluded":false},{"timestamp":5.2,"box":[0.3239583373069763,0.6722221970558167,0.36250000447034836,0.7898147851228714],"occluded":false},{"timestamp":5.46,"box":[0.35104167461395264,0.6499999761581421,0.3697916753590107,0.7657407149672508],"occluded":false},{"timestamp":5.72,"box":[0.3604166805744171,0.6638888716697693,0.38020834885537624,0.7388888671994209],"occluded":false},{"timestamp":6.5,"box":[0.33802086114883423,0.654629647731781,0.3760416954755783,0.7481481656432152],"occluded":false},{"timestamp":6.76,"box":[0.34843751788139343,0.6462962627410889,0.3651041854172945,0.7249999642372131],"occluded":false},{"timestamp":7.28,"box":[0.2958333492279053,0.645370364189148,0.3296875171363354,0.7212962880730629],"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]},{"noun_phrase":"man","object_id":2,"observations":[{"timestamp":2.08,"box":[0.3505208492279053,0.6981481313705444,0.3661458492279053,0.7898148000240326],"occluded":false},{"timestamp":2.34,"box":[0.3906250298023224,0.6888888478279114,0.4057291969656944,0.7768518105149269],"occluded":false},{"timestamp":2.6,"box":[0.4281250238418579,0.6870370507240295,0.4437500238418579,0.7750000134110451],"occluded":false},{"timestamp":2.86,"box":[0.4515625238418579,0.6787036657333374,0.477604191750288,0.7657407000660896],"occluded":false},{"timestamp":3.12,"box":[0.45364585518836975,0.6740740537643433,0.4677083557471633,0.7592592388391495],"occluded":false},{"timestamp":3.38,"box":[0.4255208671092987,0.6861110925674438,0.4390625348314643,0.7731481269001961],"occluded":false},{"timestamp":3.64,"box":[0.35885417461395264,0.6972222328186035,0.37864584289491177,0.7870370447635651],"occluded":false},{"timestamp":3.9,"box":[0.3203125298023224,0.710185170173645,0.3354166969656944,0.8027777597308159],"occluded":false},{"timestamp":4.16,"box":[0.2958333492279053,0.7074074149131775,0.3213541842997074,0.8000000044703484],"occluded":false},{"timestamp":4.42,"box":[0.2822916805744171,0.7629629373550415,0.2968750149011612,0.801851823925972],"occluded":false},{"timestamp":4.68,"box":[0.2718750238418579,0.7546296119689941,0.29375002533197403,0.787962943315506],"occluded":false},{"timestamp":4.94,"box":[0.2968750298023224,0.7462962865829468,0.31093753036111593,0.7722222115844488],"occluded":false},{"timestamp":6.5,"box":[0.3333333432674408,0.6574074029922485,0.3489583432674408,0.7481481432914734],"occluded":false},{"timestamp":6.76,"box":[0.32447919249534607,0.6499999761581421,0.3520833607763052,0.7435184940695763],"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]},{"noun_phrase":"man","object_id":3,"observations":[{"timestamp":9.36,"box":[0.0,0.2101851850748062,0.38229167461395264,0.9842592626810074],"occluded":false},{"timestamp":9.620000000000001,"box":[0.0,0.21574074029922485,0.39375001192092896,0.9907407164573669],"occluded":false},{"timestamp":9.88,"box":[0.0,0.18703703582286835,0.3973958492279053,0.9814814776182175],"occluded":false}],"scores":[0.8750000596046448,0.8750000596046448,null]},{"noun_phrase":"man","object_id":4,"observations":[{"timestamp":16.64,"box":[0.0,0.11851851642131805,0.44322919845581055,0.9712962955236435],"occluded":false},{"timestamp":16.9,"box":[0.0,0.17314814031124115,0.43020835518836975,0.9712962955236435],"occluded":false},{"timestamp":17.16,"box":[0.0,0.21481481194496155,0.43281251192092896,0.9712962806224823],"occluded":false}],"scores":[0.7460317611694336,0.7460317611694336,null]},{"noun_phrase":"man with beard","object_id":0,"observations":[{"timestamp":0.0,"box":[0.5562500357627869,0.07962962985038757,0.9994792342185974,0.9703703820705414],"occluded":false},{"timestamp":0.26,"box":[0.34322917461395264,0.20370370149612427,0.9708333611488342,0.9712962508201599],"occluded":false},{"timestamp":0.52,"box":[0.3687500059604645,0.1787036955356598,0.999479204416275,0.9722221791744232],"occluded":false},{"timestamp":0.78,"box":[0.4453125298023224,0.06666666269302368,0.999479204416275,0.9712962508201599],"occluded":false},{"timestamp":1.04,"box":[0.4515625238418579,0.10555555671453476,0.9885417222976685,0.9722222313284874],"occluded":false},{"timestamp":1.3,"box":[0.4401041865348816,0.10833333432674408,0.9708333611488342,0.9722221940755844],"occluded":false},{"timestamp":1.56,"box":[0.4609375298023224,0.04537037014961243,0.999479204416275,0.9722221791744232],"occluded":false},{"timestamp":1.82,"box":[0.4567708671092987,0.13518518209457397,0.9895834028720856,0.9722222089767456],"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":[0.4505208432674408,0.09166666865348816,0.9708333909511566,0.9722221791744232],"occluded":false},{"timestamp":2.86,"box":[0.4437500238418579,0.13611111044883728,0.9729167222976685,0.9722221791744232],"occluded":false},{"timestamp":3.12,"box":[0.44166669249534607,0.1527777761220932,0.9828125536441803,0.9722221940755844],"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":[0.4598958492279053,0.13425925374031067,0.9994792342185974,0.9722221791744232],"occluded":false},{"timestamp":4.42,"box":[0.4583333432674408,0.11481481045484543,0.999479204416275,0.972222201526165],"occluded":false},{"timestamp":4.68,"box":[0.4453125298023224,0.09074074029922485,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":4.94,"box":[0.4375000298023224,0.13148148357868195,0.999479204416275,0.9722222238779068],"occluded":false},{"timestamp":5.2,"box":[0.44739586114883423,0.09444444626569748,0.9994792342185974,0.972222201526165],"occluded":false},{"timestamp":5.46,"box":[0.4453125298023224,0.07129629701375961,0.999479204416275,0.9722222313284874],"occluded":false},{"timestamp":5.72,"box":[0.4505208432674408,0.10833333432674408,0.999479204416275,0.9722221940755844],"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":[0.44583335518836975,0.14629629254341125,0.999479204416275,0.9722221791744232],"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":[0.4598958492279053,0.08981481194496155,0.9994792342185974,0.9722221791744232],"occluded":false},{"timestamp":8.32,"box":[0.4661458432674408,0.13518518209457397,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":8.58,"box":[0.46145835518836975,0.09351851791143417,0.999479204416275,0.9712962731719017],"occluded":false},{"timestamp":8.84,"box":[0.4598958492279053,0.09351851791143417,0.9994792342185974,0.9722222313284874],"occluded":false},{"timestamp":9.1,"box":[0.4557291865348816,0.08611110597848892,0.9994792342185974,0.9722221866250038],"occluded":false},{"timestamp":9.36,"box":[0.45104169845581055,0.06666666269302368,0.9994792342185974,0.9722222089767456],"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":[0.4375000298023224,0.15925925970077515,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":10.66,"box":[0.44843751192092896,0.20000000298023224,0.9994792342185974,0.9722222238779068],"occluded":false},{"timestamp":10.92,"box":[0.4541666805744171,0.14722222089767456,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":11.18,"box":[0.43697917461395264,0.13703703880310059,0.9994791746139526,0.9722222089767456],"occluded":false},{"timestamp":11.44,"box":[0.4520833492279053,0.14166666567325592,0.9994792342185974,0.9722222238779068],"occluded":false},{"timestamp":11.700000000000001,"box":[0.44739586114883423,0.14629629254341125,0.9994792342185974,0.9722221791744232],"occluded":false},{"timestamp":11.96,"box":[0.44583335518836975,0.16944444179534912,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":12.22,"box":[0.4635416865348816,0.15555554628372192,0.9994792342185974,0.9722222089767456],"occluded":false},{"timestamp":12.48,"box":[0.4598958492279053,0.13425925374031067,0.9994792342185974,0.9722221791744232],"occluded":false},{"timestamp":12.74,"box":[0.4505208432674408,0.14074073731899261,0.999479204416275,0.9722221940755844],"occluded":false},{"timestamp":13.0,"box":[0.45468753576278687,0.1268518567085266,0.9994792342185974,0.9722222089767456],"occluded":false},{"timestamp":13.26,"box":[0.46562501788139343,0.13333332538604736,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":13.52,"box":[0.45104169845581055,0.11388888955116272,0.9994792342185974,0.9722221791744232],"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":[0.4776041805744171,0.05925925821065903,0.999479204416275,0.9722222313284874],"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":[0.4755208492279053,0.12962962687015533,0.9994792342185974,0.9722222238779068],"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":[0.45468753576278687,0.1527777761220932,0.9994792342185974,0.9722221940755844],"occluded":false},{"timestamp":16.12,"box":[0.46562501788139343,0.16203702986240387,0.999479204416275,0.9722222238779068],"occluded":false},{"timestamp":16.38,"box":[0.46562501788139343,0.15740740299224854,0.999479204416275,0.9712962508201599],"occluded":false},{"timestamp":16.64,"box":[0.45885419845581055,0.15740740299224854,0.9994792342185974,0.9722222089767456],"occluded":false},{"timestamp":16.9,"box":[0.46406251192092896,0.16018518805503845,0.9994792342185974,0.9712962806224823],"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":[0.47812503576278687,0.14999999105930328,0.9994792342185974,0.9722222238779068],"occluded":false},{"timestamp":17.94,"box":[0.47968751192092896,0.15555554628372192,0.9994792342185974,0.9712962508201599],"occluded":false},{"timestamp":18.2,"box":[0.46250003576278687,0.12222222238779068,0.9994792342185974,0.9712962880730629],"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":[0.4567708671092987,0.17222221195697784,0.999479204416275,0.9722222238779068],"occluded":false}],"scores":[0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917,0.9649122953414917]},{"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":"man","object_id":2,"presentation_id":1,"observations":[{"timestamp":2.08,"box":[0.3505208492279053,0.6981481313705444,0.3661458492279053,0.7898148000240326],"occluded":false},{"timestamp":2.34,"box":[0.3906250298023224,0.6888888478279114,0.4057291969656944,0.7768518105149269],"occluded":false},{"timestamp":2.6,"box":[0.4281250238418579,0.6870370507240295,0.4437500238418579,0.7750000134110451],"occluded":false},{"timestamp":2.86,"box":[0.4515625238418579,0.6787036657333374,0.477604191750288,0.7657407000660896],"occluded":false},{"timestamp":3.12,"box":[0.45364585518836975,0.6740740537643433,0.4677083557471633,0.7592592388391495],"occluded":false},{"timestamp":3.38,"box":[0.4255208671092987,0.6861110925674438,0.4390625348314643,0.7731481269001961],"occluded":false},{"timestamp":3.64,"box":[0.35885417461395264,0.6972222328186035,0.37864584289491177,0.7870370447635651],"occluded":false},{"timestamp":3.9,"box":[0.3203125298023224,0.710185170173645,0.3354166969656944,0.8027777597308159],"occluded":false},{"timestamp":4.16,"box":[0.2958333492279053,0.7074074149131775,0.3213541842997074,0.8000000044703484],"occluded":false},{"timestamp":4.42,"box":[0.2822916805744171,0.7629629373550415,0.2968750149011612,0.801851823925972],"occluded":false},{"timestamp":4.68,"box":[0.2718750238418579,0.7546296119689941,0.29375002533197403,0.787962943315506],"occluded":false},{"timestamp":4.94,"box":[0.2968750298023224,0.7462962865829468,0.31093753036111593,0.7722222115844488],"occluded":false},{"timestamp":6.5,"box":[0.3333333432674408,0.6574074029922485,0.3489583432674408,0.7481481432914734],"occluded":false},{"timestamp":6.76,"box":[0.32447919249534607,0.6499999761581421,0.3520833607763052,0.7435184940695763],"occluded":false}],"mean_score":0.6952381134033203},{"noun_phrase":"man","object_id":0,"presentation_id":2,"observations":[{"timestamp":0.0,"box":[0.5562500357627869,0.07962962985038757,0.9994792342185974,0.9703703820705414],"occluded":false},{"timestamp":0.26,"box":[0.34322917461395264,0.20462962985038757,0.9708333611488342,0.9722221791744232],"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":[0.4515625238418579,0.10648147761821747,0.9885417222976685,0.9722221940755844],"occluded":false},{"timestamp":1.3,"box":[0.4401041865348816,0.1101851835846901,0.9708333611488342,0.9722221866250038],"occluded":false},{"timestamp":1.56,"box":[0.4609375298023224,0.043518517166376114,0.999479204416275,0.9722221828997135],"occluded":false},{"timestamp":1.82,"box":[0.3661458492279053,0.13518518209457397,0.9895833730697632,0.9722222089767456],"occluded":false},{"timestamp":2.08,"box":[0.45625001192092896,0.1944444477558136,0.9781250357627869,0.972222238779068],"occluded":false},{"timestamp":2.34,"box":[0.39375001192092896,0.10740740597248077,0.9755208492279053,0.9722222238779068],"occluded":false},{"timestamp":2.6,"box":[0.43541669845581055,0.09259258955717087,0.970833420753479,0.972222201526165],"occluded":false},{"timestamp":2.86,"box":[0.4437500238418579,0.13518518209457397,0.9729167222976685,0.9722222089767456],"occluded":false},{"timestamp":3.12,"box":[0.44166669249534607,0.14907407760620117,0.9828125536441803,0.9722222089767456],"occluded":false},{"timestamp":3.38,"box":[0.45625001192092896,0.12222222238779068,0.9994792342185974,0.9722221866250038],"occluded":false},{"timestamp":3.64,"box":[0.45781251788139343,0.1268518567085266,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":3.9,"box":[0.45885419845581055,0.1268518567085266,0.9994792342185974,0.9722222089767456],"occluded":false},{"timestamp":4.16,"box":[0.4593750238418579,0.15833333134651184,0.9994792342185974,0.9722221791744232],"occluded":false},{"timestamp":4.42,"box":[0.45781251788139343,0.10925925523042679,0.999479204416275,0.9722222164273262],"occluded":false},{"timestamp":4.68,"box":[0.29635417461395264,0.08055555075407028,0.9994792342185974,0.972222201526165],"occluded":false},{"timestamp":4.94,"box":[0.4364583492279053,0.12962962687015533,0.9994792342185974,0.9722222238779068],"occluded":false},{"timestamp":5.2,"box":[0.31041666865348816,0.09351851791143417,0.999479204416275,0.9731481298804283],"occluded":false},{"timestamp":5.46,"box":[0.2828125059604645,0.07037036865949631,0.999479204416275,0.972222201526165],"occluded":false},{"timestamp":5.72,"box":[0.4505208432674408,0.10740740597248077,0.999479204416275,0.9722222238779068],"occluded":false},{"timestamp":5.98,"box":[0.3661458492279053,0.12962962687015533,0.9937500357627869,0.9722222238779068],"occluded":false},{"timestamp":6.24,"box":[0.24114584922790527,0.13333332538604736,0.9994792342185974,0.9990740418434143],"occluded":false},{"timestamp":6.5,"box":[0.33802086114883423,0.14722222089767456,0.9994792342185974,0.9731481075286865],"occluded":false},{"timestamp":6.76,"box":[0.33802086114883423,0.12037036567926407,0.9994792342185974,0.9722221866250038],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.3656250238418579,0.14444443583488464,0.9994792342185974,0.9722221791744232],"occluded":false},{"timestamp":7.28,"box":[0.31041666865348816,0.13425925374031067,0.999479204416275,0.9722221791744232],"occluded":false},{"timestamp":7.54,"box":[0.3098958432674408,0.0962962955236435,0.999479204416275,0.9722221940755844],"occluded":false},{"timestamp":7.8,"box":[0.29635417461395264,0.09351851791143417,0.9994792342185974,0.9722222313284874],"occluded":false},{"timestamp":8.06,"box":[0.24114584922790527,0.08796296268701553,0.9994792342185974,0.9722221866250038],"occluded":false},{"timestamp":8.32,"box":[0.4661458432674408,0.13703703880310059,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":8.58,"box":[0.46145835518836975,0.09166666865348816,0.999479204416275,0.9722221791744232],"occluded":false},{"timestamp":8.84,"box":[0.4593750238418579,0.09074074029922485,0.9994792342185974,0.9722222089767456],"occluded":false},{"timestamp":9.1,"box":[0.2828125059604645,0.07685185223817825,0.999479204416275,0.9722222164273262],"occluded":false},{"timestamp":9.36,"box":[0.4505208432674408,0.060185182839632034,0.999479204416275,0.9731481559574604],"occluded":false},{"timestamp":9.620000000000001,"box":[0.37552085518836975,0.14907407760620117,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":9.88,"box":[0.37187501788139343,0.1546296328306198,0.999479204416275,0.9722221940755844],"occluded":false},{"timestamp":10.14,"box":[0.11614584177732468,0.13703703880310059,0.9968750402331352,0.9722222089767456],"occluded":false},{"timestamp":10.4,"box":[0.03281250223517418,0.13703703880310059,0.9994792006909847,0.9722222089767456],"occluded":false},{"timestamp":10.66,"box":[0.3661458492279053,0.19907407462596893,0.9994792342185974,0.9722221940755844],"occluded":false},{"timestamp":10.92,"box":[0.32447919249534607,0.14537036418914795,0.999479204416275,0.9731481075286865],"occluded":false},{"timestamp":11.18,"box":[0.4359375238418579,0.13518518209457397,0.9994792342185974,0.9722222089767456],"occluded":false},{"timestamp":11.44,"box":[0.4520833492279053,0.14074073731899261,0.9994792342185974,0.9722221940755844],"occluded":false},{"timestamp":11.700000000000001,"box":[0.44739586114883423,0.14537036418914795,0.9994792342185974,0.9722222089767456],"occluded":false},{"timestamp":11.96,"box":[0.4453125298023224,0.1675925850868225,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":12.22,"box":[0.4635416865348816,0.1546296328306198,0.9994792342185974,0.9722221940755844],"occluded":false},{"timestamp":12.48,"box":[0.4598958492279053,0.13611111044883728,0.9994792342185974,0.9722221791744232],"occluded":false},{"timestamp":12.74,"box":[0.45000001788139343,0.1398148089647293,0.999479204416275,0.9722222238779068],"occluded":false},{"timestamp":13.0,"box":[0.45468753576278687,0.1268518567085266,0.9994792342185974,0.9722222089767456],"occluded":false},{"timestamp":13.26,"box":[0.46562501788139343,0.13055555522441864,0.999479204416275,0.9722221940755844],"occluded":false},{"timestamp":13.52,"box":[0.4505208432674408,0.11203703284263611,0.999479204416275,0.9722221791744232],"occluded":false},{"timestamp":13.780000000000001,"box":[0.3765625059604645,0.07037036865949631,0.999479204416275,0.972222201526165],"occluded":false},{"timestamp":14.040000000000001,"box":[0.35104167461395264,0.06481481343507767,0.9994792342185974,0.9722222164273262],"occluded":false},{"timestamp":14.3,"box":[0.21302084624767303,0.05925925821065903,0.9994792193174362,0.9722222313284874],"occluded":false},{"timestamp":14.56,"box":[0.4817708730697632,0.08148147910833359,0.9994792342185974,0.9722222313284874],"occluded":false},{"timestamp":14.82,"box":[0.4833333492279053,0.10092592239379883,0.9994792342185974,0.9722222089767456],"occluded":false},{"timestamp":15.08,"box":[0.4755208492279053,0.12962962687015533,0.9994792342185974,0.9722222238779068],"occluded":false},{"timestamp":15.34,"box":[0.452604204416275,0.14351852238178253,0.999479204416275,0.9722222238779068],"occluded":false},{"timestamp":15.6,"box":[0.06041666865348816,0.15648147463798523,0.999479204416275,0.9722221791744232],"occluded":false},{"timestamp":15.860000000000001,"box":[0.45468753576278687,0.1509259194135666,0.9994792342185974,0.9722221940755844],"occluded":false},{"timestamp":16.12,"box":[0.13333334028720856,0.16203702986240387,0.9994791895151138,0.9722222238779068],"occluded":false},{"timestamp":16.38,"box":[0.46562501788139343,0.15648147463798523,0.999479204416275,0.9722221791744232],"occluded":false},{"timestamp":16.64,"box":[0.4593750238418579,0.15833333134651184,0.9994792342185974,0.9722221791744232],"occluded":false},{"timestamp":16.9,"box":[0.46406251192092896,0.15925925970077515,0.9994792342185974,0.9712963104248047],"occluded":false},{"timestamp":17.16,"box":[0.46927085518836975,0.16296295821666718,0.999479204416275,0.9722221940755844],"occluded":false},{"timestamp":17.42,"box":[0.4661458432674408,0.1546296328306198,0.999479204416275,0.9722221940755844],"occluded":false},{"timestamp":17.68,"box":[0.47708335518836975,0.14907407760620117,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":17.94,"box":[0.2828125059604645,0.1537037044763565,0.999479204416275,0.9722222238779068],"occluded":false},{"timestamp":18.2,"box":[0.07447917014360428,0.12129629403352737,0.999479241669178,0.9722222164273262],"occluded":false},{"timestamp":18.46,"box":[0.03281250223517418,0.03611110895872116,0.9994792006909847,0.9712963029742241],"occluded":false},{"timestamp":18.72,"box":[0.10208334028720856,0.031481482088565826,0.9994792491197586,0.9722221866250038],"occluded":false},{"timestamp":18.98,"box":[0.4286458492279053,0.03333333134651184,0.9994792342185974,0.9712962806224823],"occluded":false},{"timestamp":19.240000000000002,"box":[0.43177086114883423,0.09259258955717087,0.9994792342185974,0.972222201526165],"occluded":false},{"timestamp":19.5,"box":[0.4359375238418579,0.1398148089647293,0.9994792342185974,0.9722222238779068],"occluded":false},{"timestamp":19.76,"box":[0.43281251192092896,0.18333333730697632,0.9994792342185974,0.9722222089767456],"occluded":false},{"timestamp":20.02,"box":[0.4442708492279053,0.18796296417713165,0.9994792342185974,0.9722222238779068],"occluded":false},{"timestamp":20.28,"box":[0.4567708671092987,0.17037037014961243,0.999479204416275,0.9722221791744232],"occluded":false}],"mean_score":0.949367105960846},{"noun_phrase":"man","object_id":1,"presentation_id":3,"observations":[{"timestamp":1.3,"box":[0.4046875238418579,0.6564814448356628,0.425000024959445,0.758333295583725],"occluded":false},{"timestamp":1.56,"box":[0.3593750298023224,0.6879629492759705,0.38072919845581055,0.7925925776362419],"occluded":false},{"timestamp":1.82,"box":[0.3541666865348816,0.6814814805984497,0.3744791876524687,0.7870370373129845],"occluded":false},{"timestamp":2.08,"box":[0.37916669249534607,0.690740704536438,0.4000000264495611,0.7953703328967094],"occluded":false},{"timestamp":2.34,"box":[0.41875001788139343,0.6833333373069763,0.4395833518356085,0.7861111164093018],"occluded":false},{"timestamp":2.6,"box":[0.45625001192092896,0.6824073791503906,0.475520845502615,0.7842592298984528],"occluded":false},{"timestamp":3.38,"box":[0.4541666805744171,0.6796296238899231,0.47395834885537624,0.7787036970257759],"occluded":false},{"timestamp":3.64,"box":[0.3895833492279053,0.6916666626930237,0.4114583507180214,0.7944444417953491],"occluded":false},{"timestamp":3.9,"box":[0.3500000238418579,0.7037037014961243,0.37343752570450306,0.8101851791143417],"occluded":false},{"timestamp":4.16,"box":[0.3270833492279053,0.7018518447875977,0.34947918355464935,0.8092592507600784],"occluded":false},{"timestamp":4.42,"box":[0.3109375238418579,0.7092592716217041,0.334895858541131,0.8194444552063942],"occluded":false},{"timestamp":4.68,"box":[0.3062500059604645,0.6953703761100769,0.32864584028720856,0.8046296313405037],"occluded":false},{"timestamp":4.94,"box":[0.32864585518836975,0.6796296238899231,0.3520833570510149,0.7851851806044579],"occluded":false},{"timestamp":5.2,"box":[0.3239583373069763,0.6722221970558167,0.36250000447034836,0.7898147851228714],"occluded":false},{"timestamp":5.46,"box":[0.35104167461395264,0.6499999761581421,0.3697916753590107,0.7657407149672508],"occluded":false},{"timestamp":5.72,"box":[0.3604166805744171,0.6638888716697693,0.38020834885537624,0.7388888671994209],"occluded":false},{"timestamp":6.5,"box":[0.33802086114883423,0.654629647731781,0.3760416954755783,0.7481481656432152],"occluded":false},{"timestamp":6.76,"box":[0.34843751788139343,0.6462962627410889,0.3651041854172945,0.7249999642372131],"occluded":false},{"timestamp":7.28,"box":[0.2958333492279053,0.645370364189148,0.3296875171363354,0.7212962880730629],"occluded":false}],"mean_score":0.7697842121124268},{"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":{"merged_duplicate":4},"fates":[{"noun_phrase":"man","object_id":0,"frames":79,"mean_score":0.949367105960846,"mean_area":0.5233432658423902,"outcome":"kept","presentation_id":2,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"man","object_id":1,"frames":19,"mean_score":0.7697842121124268,"mean_area":0.002369817127404059,"outcome":"kept","presentation_id":3,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"man","object_id":2,"frames":14,"mean_score":0.6952381134033203,"mean_area":0.001426986919635822,"outcome":"kept","presentation_id":1,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"man","object_id":3,"frames":3,"mean_score":0.8750000596046448,"mean_area":0.3055957496219885,"outcome":"merged_duplicate","presentation_id":null,"merged_into_phrase":"woman","merged_into_object_id":0,"merge_iou":0.9724600993386711},{"noun_phrase":"man","object_id":4,"frames":3,"mean_score":0.7460317611694336,"mean_area":0.3495868870743202,"outcome":"merged_duplicate","presentation_id":null,"merged_into_phrase":"woman","merged_into_object_id":0,"merge_iou":0.8642530158294349},{"noun_phrase":"man with beard","object_id":0,"frames":79,"mean_score":0.9649122953414917,"mean_area":0.4662899150846834,"outcome":"merged_duplicate","presentation_id":null,"merged_into_phrase":"man","merged_into_object_id":0,"merge_iou":0.9146618825123337},{"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":"merged_duplicate","presentation_id":null,"merged_into_phrase":"man","merged_into_object_id":1,"merge_iou":0.9171681865404692}],"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/tmp/rmot_smoke_real_86260/renders/--cif7Nv6BI_chunk_015_scene_004_choices.mp4","media_bytes":7159836,"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","temperature":0.0,"as_video":true,"media_bytes":7159836,"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 dreadlocks and a beard\n2. the woman who raises her hand to shield her eyes from the sun\n3. the man wearing a blue cap","questions":["the man with dreadlocks and a beard","the woman who raises her hand to shield her eyes from the sun","the man wearing a blue cap"]},"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\": 7,\n      \"end_frame\": 21,\n      \"confidence\": 1.0\n    },\n    {\n      \"description_index\": 3,\n      \"matching_ids\": [\n        3\n      ],\n      \"start_frame\": 3,\n      \"end_frame\": 6,\n      \"confidence\": 0.8\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":7,"end_frame":21,"confidence":1.0},{"description_index":3,"matching_ids":[3],"start_frame":3,"end_frame":6,"confidence":0.8}],"answers":[{"expression":"the man with dreadlocks and a beard","matching_presentation_ids":[2],"start_choice_frame":1,"end_choice_frame":21,"confidence":1.0,"self_reported_confidence":1.0},{"expression":"the woman who raises her hand to shield her eyes from the sun","matching_presentation_ids":[4],"start_choice_frame":7,"end_choice_frame":21,"confidence":1.0,"self_reported_confidence":1.0},{"expression":"the man wearing a blue cap","matching_presentation_ids":[3],"start_choice_frame":3,"end_choice_frame":6,"confidence":0.8,"self_reported_confidence":0.8}]}},"verify":{"inputs":{"expressions":3,"tracklets":4,"selections":3,"gates":"interval, count, unknown ids, confidence, then geometry (motion)"},"outputs":{"kept":[{"expression":"the man with dreadlocks and a beard","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":[[0.5562500357627869,0.07962962985038757,0.9994792342185974,0.9703703820705414],[0.34322917461395264,0.20462962985038757,0.9708333611488342,0.9722221791744232],[0.3687500059604645,0.1787036955356598,0.999479204416275,0.9722221791744232],[0.4453125298023224,0.06759259104728699,0.999479204416275,0.9712962806224823],[0.4515625238418579,0.10648147761821747,0.9885417222976685,0.9722221940755844],[0.4401041865348816,0.1101851835846901,0.9708333611488342,0.9722221866250038],[0.4609375298023224,0.043518517166376114,0.999479204416275,0.9722221828997135],[0.3661458492279053,0.13518518209457397,0.9895833730697632,0.9722222089767456],[0.45625001192092896,0.1944444477558136,0.9781250357627869,0.972222238779068],[0.39375001192092896,0.10740740597248077,0.9755208492279053,0.9722222238779068],[0.43541669845581055,0.09259258955717087,0.970833420753479,0.972222201526165],[0.4437500238418579,0.13518518209457397,0.9729167222976685,0.9722222089767456],[0.44166669249534607,0.14907407760620117,0.9828125536441803,0.9722222089767456],[0.45625001192092896,0.12222222238779068,0.9994792342185974,0.9722221866250038],[0.45781251788139343,0.1268518567085266,0.999479204416275,0.9722222089767456],[0.45885419845581055,0.1268518567085266,0.9994792342185974,0.9722222089767456],[0.4593750238418579,0.15833333134651184,0.9994792342185974,0.9722221791744232],[0.45781251788139343,0.10925925523042679,0.999479204416275,0.9722222164273262],[0.29635417461395264,0.08055555075407028,0.9994792342185974,0.972222201526165],[0.4364583492279053,0.12962962687015533,0.9994792342185974,0.9722222238779068],[0.31041666865348816,0.09351851791143417,0.999479204416275,0.9731481298804283],[0.2828125059604645,0.07037036865949631,0.999479204416275,0.972222201526165],[0.4505208432674408,0.10740740597248077,0.999479204416275,0.9722222238779068],[0.3661458492279053,0.12962962687015533,0.9937500357627869,0.9722222238779068],[0.24114584922790527,0.13333332538604736,0.9994792342185974,0.9990740418434143],[0.33802086114883423,0.14722222089767456,0.9994792342185974,0.9731481075286865],[0.33802086114883423,0.12037036567926407,0.9994792342185974,0.9722221866250038],[0.3656250238418579,0.14444443583488464,0.9994792342185974,0.9722221791744232],[0.31041666865348816,0.13425925374031067,0.999479204416275,0.9722221791744232],[0.3098958432674408,0.0962962955236435,0.999479204416275,0.9722221940755844],[0.29635417461395264,0.09351851791143417,0.9994792342185974,0.9722222313284874],[0.24114584922790527,0.08796296268701553,0.9994792342185974,0.9722221866250038],[0.4661458432674408,0.13703703880310059,0.999479204416275,0.9722222089767456],[0.46145835518836975,0.09166666865348816,0.999479204416275,0.9722221791744232],[0.4593750238418579,0.09074074029922485,0.9994792342185974,0.9722222089767456],[0.2828125059604645,0.07685185223817825,0.999479204416275,0.9722222164273262],[0.4505208432674408,0.060185182839632034,0.999479204416275,0.9731481559574604],[0.37552085518836975,0.14907407760620117,0.999479204416275,0.9722222089767456],[0.37187501788139343,0.1546296328306198,0.999479204416275,0.9722221940755844],[0.11614584177732468,0.13703703880310059,0.9968750402331352,0.9722222089767456],[0.03281250223517418,0.13703703880310059,0.9994792006909847,0.9722222089767456],[0.3661458492279053,0.19907407462596893,0.9994792342185974,0.9722221940755844],[0.32447919249534607,0.14537036418914795,0.999479204416275,0.9731481075286865],[0.4359375238418579,0.13518518209457397,0.9994792342185974,0.9722222089767456],[0.4520833492279053,0.14074073731899261,0.9994792342185974,0.9722221940755844],[0.44739586114883423,0.14537036418914795,0.9994792342185974,0.9722222089767456],[0.4453125298023224,0.1675925850868225,0.999479204416275,0.9722222089767456],[0.4635416865348816,0.1546296328306198,0.9994792342185974,0.9722221940755844],[0.4598958492279053,0.13611111044883728,0.9994792342185974,0.9722221791744232],[0.45000001788139343,0.1398148089647293,0.999479204416275,0.9722222238779068],[0.45468753576278687,0.1268518567085266,0.9994792342185974,0.9722222089767456],[0.46562501788139343,0.13055555522441864,0.999479204416275,0.9722221940755844],[0.4505208432674408,0.11203703284263611,0.999479204416275,0.9722221791744232],[0.3765625059604645,0.07037036865949631,0.999479204416275,0.972222201526165],[0.35104167461395264,0.06481481343507767,0.9994792342185974,0.9722222164273262],[0.21302084624767303,0.05925925821065903,0.9994792193174362,0.9722222313284874],[0.4817708730697632,0.08148147910833359,0.9994792342185974,0.9722222313284874],[0.4833333492279053,0.10092592239379883,0.9994792342185974,0.9722222089767456],[0.4755208492279053,0.12962962687015533,0.9994792342185974,0.9722222238779068],[0.452604204416275,0.14351852238178253,0.999479204416275,0.9722222238779068],[0.06041666865348816,0.15648147463798523,0.999479204416275,0.9722221791744232],[0.45468753576278687,0.1509259194135666,0.9994792342185974,0.9722221940755844],[0.13333334028720856,0.16203702986240387,0.9994791895151138,0.9722222238779068],[0.46562501788139343,0.15648147463798523,0.999479204416275,0.9722221791744232],[0.4593750238418579,0.15833333134651184,0.9994792342185974,0.9722221791744232],[0.46406251192092896,0.15925925970077515,0.9994792342185974,0.9712963104248047],[0.46927085518836975,0.16296295821666718,0.999479204416275,0.9722221940755844],[0.4661458432674408,0.1546296328306198,0.999479204416275,0.9722221940755844],[0.47708335518836975,0.14907407760620117,0.999479204416275,0.9722222089767456],[0.2828125059604645,0.1537037044763565,0.999479204416275,0.9722222238779068],[0.07447917014360428,0.12129629403352737,0.999479241669178,0.9722222164273262],[0.03281250223517418,0.03611110895872116,0.9994792006909847,0.9712963029742241],[0.10208334028720856,0.031481482088565826,0.9994792491197586,0.9722221866250038],[0.4286458492279053,0.03333333134651184,0.9994792342185974,0.9712962806224823],[0.43177086114883423,0.09259258955717087,0.9994792342185974,0.972222201526165],[0.4359375238418579,0.1398148089647293,0.9994792342185974,0.9722222238779068],[0.43281251192092896,0.18333333730697632,0.9994792342185974,0.9722222089767456],[0.4442708492279053,0.18796296417713165,0.9994792342185974,0.9722222238779068],[0.4567708671092987,0.17037037014961243,0.999479204416275,0.9722221791744232]]}]},{"expression":"the woman who raises her hand to shield her eyes from the sun","kind":"motion","is_negative":false,"tracks":[{"timestamps":[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":[[0.0,0.3018518388271332,0.41822919249534607,0.9712962806224823],[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]]}]}],"rejected":{"the man wearing a blue cap":"selection_matched"},"rejection_counts":{"selection_matched":1},"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 dreadlocks and a beard","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":[[0.5562500357627869,0.07962962985038757,0.9994792342185974,0.9703703820705414],[0.34322917461395264,0.20462962985038757,0.9708333611488342,0.9722221791744232],[0.3687500059604645,0.1787036955356598,0.999479204416275,0.9722221791744232],[0.4453125298023224,0.06759259104728699,0.999479204416275,0.9712962806224823],[0.4515625238418579,0.10648147761821747,0.9885417222976685,0.9722221940755844],[0.4401041865348816,0.1101851835846901,0.9708333611488342,0.9722221866250038],[0.4609375298023224,0.043518517166376114,0.999479204416275,0.9722221828997135],[0.3661458492279053,0.13518518209457397,0.9895833730697632,0.9722222089767456],[0.45625001192092896,0.1944444477558136,0.9781250357627869,0.972222238779068],[0.39375001192092896,0.10740740597248077,0.9755208492279053,0.9722222238779068],[0.43541669845581055,0.09259258955717087,0.970833420753479,0.972222201526165],[0.4437500238418579,0.13518518209457397,0.9729167222976685,0.9722222089767456],[0.44166669249534607,0.14907407760620117,0.9828125536441803,0.9722222089767456],[0.45625001192092896,0.12222222238779068,0.9994792342185974,0.9722221866250038],[0.45781251788139343,0.1268518567085266,0.999479204416275,0.9722222089767456],[0.45885419845581055,0.1268518567085266,0.9994792342185974,0.9722222089767456],[0.4593750238418579,0.15833333134651184,0.9994792342185974,0.9722221791744232],[0.45781251788139343,0.10925925523042679,0.999479204416275,0.9722222164273262],[0.29635417461395264,0.08055555075407028,0.9994792342185974,0.972222201526165],[0.4364583492279053,0.12962962687015533,0.9994792342185974,0.9722222238779068],[0.31041666865348816,0.09351851791143417,0.999479204416275,0.9731481298804283],[0.2828125059604645,0.07037036865949631,0.999479204416275,0.972222201526165],[0.4505208432674408,0.10740740597248077,0.999479204416275,0.9722222238779068],[0.3661458492279053,0.12962962687015533,0.9937500357627869,0.9722222238779068],[0.24114584922790527,0.13333332538604736,0.9994792342185974,0.9990740418434143],[0.33802086114883423,0.14722222089767456,0.9994792342185974,0.9731481075286865],[0.33802086114883423,0.12037036567926407,0.9994792342185974,0.9722221866250038],[0.3656250238418579,0.14444443583488464,0.9994792342185974,0.9722221791744232],[0.31041666865348816,0.13425925374031067,0.999479204416275,0.9722221791744232],[0.3098958432674408,0.0962962955236435,0.999479204416275,0.9722221940755844],[0.29635417461395264,0.09351851791143417,0.9994792342185974,0.9722222313284874],[0.24114584922790527,0.08796296268701553,0.9994792342185974,0.9722221866250038],[0.4661458432674408,0.13703703880310059,0.999479204416275,0.9722222089767456],[0.46145835518836975,0.09166666865348816,0.999479204416275,0.9722221791744232],[0.4593750238418579,0.09074074029922485,0.9994792342185974,0.9722222089767456],[0.2828125059604645,0.07685185223817825,0.999479204416275,0.9722222164273262],[0.4505208432674408,0.060185182839632034,0.999479204416275,0.9731481559574604],[0.37552085518836975,0.14907407760620117,0.999479204416275,0.9722222089767456],[0.37187501788139343,0.1546296328306198,0.999479204416275,0.9722221940755844],[0.11614584177732468,0.13703703880310059,0.9968750402331352,0.9722222089767456],[0.03281250223517418,0.13703703880310059,0.9994792006909847,0.9722222089767456],[0.3661458492279053,0.19907407462596893,0.9994792342185974,0.9722221940755844],[0.32447919249534607,0.14537036418914795,0.999479204416275,0.9731481075286865],[0.4359375238418579,0.13518518209457397,0.9994792342185974,0.9722222089767456],[0.4520833492279053,0.14074073731899261,0.9994792342185974,0.9722221940755844],[0.44739586114883423,0.14537036418914795,0.9994792342185974,0.9722222089767456],[0.4453125298023224,0.1675925850868225,0.999479204416275,0.9722222089767456],[0.4635416865348816,0.1546296328306198,0.9994792342185974,0.9722221940755844],[0.4598958492279053,0.13611111044883728,0.9994792342185974,0.9722221791744232],[0.45000001788139343,0.1398148089647293,0.999479204416275,0.9722222238779068],[0.45468753576278687,0.1268518567085266,0.9994792342185974,0.9722222089767456],[0.46562501788139343,0.13055555522441864,0.999479204416275,0.9722221940755844],[0.4505208432674408,0.11203703284263611,0.999479204416275,0.9722221791744232],[0.3765625059604645,0.07037036865949631,0.999479204416275,0.972222201526165],[0.35104167461395264,0.06481481343507767,0.9994792342185974,0.9722222164273262],[0.21302084624767303,0.05925925821065903,0.9994792193174362,0.9722222313284874],[0.4817708730697632,0.08148147910833359,0.9994792342185974,0.9722222313284874],[0.4833333492279053,0.10092592239379883,0.9994792342185974,0.9722222089767456],[0.4755208492279053,0.12962962687015533,0.9994792342185974,0.9722222238779068],[0.452604204416275,0.14351852238178253,0.999479204416275,0.9722222238779068],[0.06041666865348816,0.15648147463798523,0.999479204416275,0.9722221791744232],[0.45468753576278687,0.1509259194135666,0.9994792342185974,0.9722221940755844],[0.13333334028720856,0.16203702986240387,0.9994791895151138,0.9722222238779068],[0.46562501788139343,0.15648147463798523,0.999479204416275,0.9722221791744232],[0.4593750238418579,0.15833333134651184,0.9994792342185974,0.9722221791744232],[0.46406251192092896,0.15925925970077515,0.9994792342185974,0.9712963104248047],[0.46927085518836975,0.16296295821666718,0.999479204416275,0.9722221940755844],[0.4661458432674408,0.1546296328306198,0.999479204416275,0.9722221940755844],[0.47708335518836975,0.14907407760620117,0.999479204416275,0.9722222089767456],[0.2828125059604645,0.1537037044763565,0.999479204416275,0.9722222238779068],[0.07447917014360428,0.12129629403352737,0.999479241669178,0.9722222164273262],[0.03281250223517418,0.03611110895872116,0.9994792006909847,0.9712963029742241],[0.10208334028720856,0.031481482088565826,0.9994792491197586,0.9722221866250038],[0.4286458492279053,0.03333333134651184,0.9994792342185974,0.9712962806224823],[0.43177086114883423,0.09259258955717087,0.9994792342185974,0.972222201526165],[0.4359375238418579,0.1398148089647293,0.9994792342185974,0.9722222238779068],[0.43281251192092896,0.18333333730697632,0.9994792342185974,0.9722222089767456],[0.4442708492279053,0.18796296417713165,0.9994792342185974,0.9722222238779068],[0.4567708671092987,0.17037037014961243,0.999479204416275,0.9722221791744232]]}]},{"expression":"the woman who raises her hand to shield her eyes from the sun","kind":"motion","is_negative":false,"tracks":[{"timestamps":[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":[[0.0,0.3018518388271332,0.41822919249534607,0.9712962806224823],[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]]}]}],"rmot_expression_count":2,"rmot_negative_count":0,"rmot_motion_count":1}}}}},{"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","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 massive boulder resting on top of the building with the red door\",\n      \"noun_phrase\": \"massive boulder\",\n      \"kind\": \"spatial\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the stone building with a red door underneath the giant boulder\",\n      \"noun_phrase\": \"stone building with red door\",\n      \"kind\": \"spatial\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the black car driving down the street\",\n      \"noun_phrase\": \"black car\",\n      \"kind\": \"motion\",\n      \"expected_count\": 0,\n      \"is_negative\": true\n    }\n  ]\n}","raw_items":[{"expression":"the massive boulder resting on top of the building with the red door","noun_phrase":"massive boulder","kind":"spatial","expected_count":1,"is_negative":false},{"expression":"the stone building with a red door underneath the giant boulder","noun_phrase":"stone building with red door","kind":"spatial","expected_count":1,"is_negative":false},{"expression":"the black car driving down the street","noun_phrase":"black car","kind":"motion","expected_count":0,"is_negative":true}],"accepted":[{"expression":"the massive boulder resting on top of the building with the red door","noun_phrase":"massive boulder","kind":"spatial","expected_count":1,"is_negative":false},{"expression":"the stone building with a red door underneath the giant boulder","noun_phrase":"stone building with red door","kind":"spatial","expected_count":1,"is_negative":false},{"expression":"the black car driving down the street","noun_phrase":"black car","kind":"motion","expected_count":0,"is_negative":true}],"gated":[]}},"track":{"inputs":{"tracker":"Sam31MultiplexTracker","model_id":"/mnt/ceph3/ec/jacky/sam3.1","noun_phrases":["black car","massive boulder","stone building with 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":"massive 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":[0.6239583492279053,0.7111110687255859,0.7687500268220901,0.9305555075407028],"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":[0.6395833492279053,0.8888888955116272,0.8104166984558105,0.9722222238779068],"occluded":false},{"timestamp":1.82,"box":[0.6406250596046448,0.8916666507720947,0.8145834058523178,0.972222201526165],"occluded":false},{"timestamp":2.08,"box":[0.6364583373069763,0.8740740418434143,0.8140625059604645,0.9722221866250038],"occluded":false},{"timestamp":2.34,"box":[0.6265625357627869,0.8287037014961243,0.8119792193174362,0.9722222238779068],"occluded":false},{"timestamp":2.6,"box":[0.6250000596046448,0.7990740537643433,0.8083333969116211,0.9722221940755844],"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":[0.6000000238418579,0.690740704536438,0.757291704416275,0.9222221821546555],"occluded":false},{"timestamp":4.16,"box":[0.6031250357627869,0.6814814805984497,0.7479167133569717,0.9009259194135666],"occluded":false},{"timestamp":4.42,"box":[0.6010416746139526,0.6787036657333374,0.7369791865348816,0.8842592239379883],"occluded":false},{"timestamp":4.68,"box":[0.5984375476837158,0.6712962985038757,0.7286458909511566,0.8694444447755814],"occluded":false},{"timestamp":4.94,"box":[0.5953125357627869,0.6675925850868225,0.7281250357627869,0.8685185015201569],"occluded":false},{"timestamp":5.2,"box":[0.5937500596046448,0.6712962985038757,0.7260417342185974,0.8694444447755814],"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":[0.7333333492279053,0.675000011920929,0.8552083522081375,0.8351851999759674],"occluded":false},{"timestamp":8.58,"box":[0.7744792103767395,0.680555522441864,0.8989583849906921,0.8370369970798492],"occluded":false},{"timestamp":8.84,"box":[0.8197917342185974,0.6870370507240295,0.9526042342185974,0.8444444537162781],"occluded":false},{"timestamp":9.1,"box":[0.8682292103767395,0.6990740895271301,0.9994792193174362,0.8694444596767426],"occluded":false},{"timestamp":9.36,"box":[0.9192708730697632,0.7249999642372131,0.9994792118668556,0.9055555164813995],"occluded":false},{"timestamp":9.620000000000001,"box":[0.96770840883255,0.7731481194496155,0.9994792453944683,0.9277777522802353],"occluded":false}],"scores":[0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781]},{"noun_phrase":"massive 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":[0.22708334028720856,0.6287037134170532,0.46822918951511383,0.9648148119449615],"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":[0.31666669249534607,0.6296296119689941,0.4854166954755783,0.8675925731658936],"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":[0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115,0.9139785170555115]},{"noun_phrase":"massive boulder","object_id":2,"observations":[{"timestamp":0.0,"box":[0.3489583432674408,0.12037036567926407,0.7291666865348816,0.6268518343567848],"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":[0.25468751788139343,0.14814814925193787,0.7421875298023224,0.8009259402751923],"occluded":false},{"timestamp":1.82,"box":[0.2562500238418579,0.14999999105930328,0.7427083849906921,0.7999999672174454],"occluded":false},{"timestamp":2.08,"box":[0.2593750059604645,0.13148148357868195,0.7468750178813934,0.7805555611848831],"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":[0.2588541805744171,0.031481482088565826,0.749479204416275,0.6675925627350807],"occluded":false},{"timestamp":3.12,"box":[0.2614583373069763,0.031481482088565826,0.7463541924953461,0.6416666284203529],"occluded":false},{"timestamp":3.38,"box":[0.27031251788139343,0.031481482088565826,0.7390625476837158,0.6194444075226784],"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":[0.3192708492279053,0.0962962955236435,0.6984375417232513,0.6018518656492233],"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":[0.32343751192092896,0.11759258806705475,0.6848958730697632,0.5999999791383743],"occluded":false},{"timestamp":5.46,"box":[0.3218750059604645,0.1240740716457367,0.6822916865348816,0.6027777791023254],"occluded":false},{"timestamp":5.72,"box":[0.31979167461395264,0.13148148357868195,0.6802083551883698,0.6129629462957382],"occluded":false},{"timestamp":5.98,"box":[0.31979167461395264,0.14907407760620117,0.6786458492279053,0.6268518567085266],"occluded":false},{"timestamp":6.24,"box":[0.31979167461395264,0.16388888657093048,0.6786458492279053,0.6407407373189926],"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":[0.49531251192092896,0.2203703671693802,0.7890625298023224,0.5981481522321701],"occluded":false},{"timestamp":8.32,"box":[0.526562511920929,0.2342592477798462,0.8161458671092987,0.6009259223937988],"occluded":false},{"timestamp":8.58,"box":[0.5677083730697632,0.23981481790542603,0.861979216337204,0.6018518507480621],"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":[0.6791666746139526,0.2361111044883728,0.999479204416275,0.6212962865829468],"occluded":false},{"timestamp":9.620000000000001,"box":[0.7098958492279053,0.2361111044883728,0.999479204416275,0.6361111104488373],"occluded":false},{"timestamp":9.88,"box":[0.7364583611488342,0.2361111044883728,0.999479204416275,0.652777761220932],"occluded":false},{"timestamp":10.14,"box":[0.7812500596046448,0.23148147761821747,0.9994792342185974,0.6768518537282944],"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":[0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333,0.9569892287254333]},{"noun_phrase":"massive boulder","object_id":3,"observations":[{"timestamp":11.18,"box":[0.0,0.6851851940155029,0.1145833358168602,0.9731481373310089],"occluded":false},{"timestamp":11.44,"box":[0.0,0.6898148059844971,0.1770833432674408,0.9759259223937988],"occluded":false},{"timestamp":11.700000000000001,"box":[0.0,0.6722221970558167,0.22500000894069672,0.9740740358829498],"occluded":false},{"timestamp":11.96,"box":[0.0,0.6675925850868225,0.2692708373069763,0.9990740716457367],"occluded":false},{"timestamp":12.22,"box":[0.004166666883975267,0.654629647731781,0.9354167026467621,0.9990740716457367],"occluded":false},{"timestamp":12.48,"box":[0.08802083879709244,0.6361110806465149,0.33906251937150955,0.9638888537883759],"occluded":false},{"timestamp":12.74,"box":[0.11562500894069672,0.6203703880310059,0.35625001788139343,0.9407407641410828],"occluded":false}],"scores":[0.7387754917144775,0.7387754917144775,0.7387754917144775,0.7387754917144775,0.7387754917144775,0.7387754917144775,0.7387754917144775]}],"tracklets":[{"noun_phrase":"massive boulder","object_id":2,"presentation_id":1,"observations":[{"timestamp":0.0,"box":[0.3489583432674408,0.12037036567926407,0.7291666865348816,0.6268518343567848],"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":[0.25468751788139343,0.14814814925193787,0.7421875298023224,0.8009259402751923],"occluded":false},{"timestamp":1.82,"box":[0.2562500238418579,0.14999999105930328,0.7427083849906921,0.7999999672174454],"occluded":false},{"timestamp":2.08,"box":[0.2593750059604645,0.13148148357868195,0.7468750178813934,0.7805555611848831],"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":[0.2588541805744171,0.031481482088565826,0.749479204416275,0.6675925627350807],"occluded":false},{"timestamp":3.12,"box":[0.2614583373069763,0.031481482088565826,0.7463541924953461,0.6416666284203529],"occluded":false},{"timestamp":3.38,"box":[0.27031251788139343,0.031481482088565826,0.7390625476837158,0.6194444075226784],"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":[0.3192708492279053,0.0962962955236435,0.6984375417232513,0.6018518656492233],"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":[0.32343751192092896,0.11759258806705475,0.6848958730697632,0.5999999791383743],"occluded":false},{"timestamp":5.46,"box":[0.3218750059604645,0.1240740716457367,0.6822916865348816,0.6027777791023254],"occluded":false},{"timestamp":5.72,"box":[0.31979167461395264,0.13148148357868195,0.6802083551883698,0.6129629462957382],"occluded":false},{"timestamp":5.98,"box":[0.31979167461395264,0.14907407760620117,0.6786458492279053,0.6268518567085266],"occluded":false},{"timestamp":6.24,"box":[0.31979167461395264,0.16388888657093048,0.6786458492279053,0.6407407373189926],"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":[0.49531251192092896,0.2203703671693802,0.7890625298023224,0.5981481522321701],"occluded":false},{"timestamp":8.32,"box":[0.526562511920929,0.2342592477798462,0.8161458671092987,0.6009259223937988],"occluded":false},{"timestamp":8.58,"box":[0.5677083730697632,0.23981481790542603,0.861979216337204,0.6018518507480621],"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":[0.6791666746139526,0.2361111044883728,0.999479204416275,0.6212962865829468],"occluded":false},{"timestamp":9.620000000000001,"box":[0.7098958492279053,0.2361111044883728,0.999479204416275,0.6361111104488373],"occluded":false},{"timestamp":9.88,"box":[0.7364583611488342,0.2361111044883728,0.999479204416275,0.652777761220932],"occluded":false},{"timestamp":10.14,"box":[0.7812500596046448,0.23148147761821747,0.9994792342185974,0.6768518537282944],"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":0.9569892287254333},{"noun_phrase":"massive boulder","object_id":0,"presentation_id":2,"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":[0.6239583492279053,0.7111110687255859,0.7687500268220901,0.9305555075407028],"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":[0.6395833492279053,0.8888888955116272,0.8104166984558105,0.9722222238779068],"occluded":false},{"timestamp":1.82,"box":[0.6406250596046448,0.8916666507720947,0.8145834058523178,0.972222201526165],"occluded":false},{"timestamp":2.08,"box":[0.6364583373069763,0.8740740418434143,0.8140625059604645,0.9722221866250038],"occluded":false},{"timestamp":2.34,"box":[0.6265625357627869,0.8287037014961243,0.8119792193174362,0.9722222238779068],"occluded":false},{"timestamp":2.6,"box":[0.6250000596046448,0.7990740537643433,0.8083333969116211,0.9722221940755844],"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":[0.6000000238418579,0.690740704536438,0.757291704416275,0.9222221821546555],"occluded":false},{"timestamp":4.16,"box":[0.6031250357627869,0.6814814805984497,0.7479167133569717,0.9009259194135666],"occluded":false},{"timestamp":4.42,"box":[0.6010416746139526,0.6787036657333374,0.7369791865348816,0.8842592239379883],"occluded":false},{"timestamp":4.68,"box":[0.5984375476837158,0.6712962985038757,0.7286458909511566,0.8694444447755814],"occluded":false},{"timestamp":4.94,"box":[0.5953125357627869,0.6675925850868225,0.7281250357627869,0.8685185015201569],"occluded":false},{"timestamp":5.2,"box":[0.5937500596046448,0.6712962985038757,0.7260417342185974,0.8694444447755814],"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":[0.7333333492279053,0.675000011920929,0.8552083522081375,0.8351851999759674],"occluded":false},{"timestamp":8.58,"box":[0.7744792103767395,0.680555522441864,0.8989583849906921,0.8370369970798492],"occluded":false},{"timestamp":8.84,"box":[0.8197917342185974,0.6870370507240295,0.9526042342185974,0.8444444537162781],"occluded":false},{"timestamp":9.1,"box":[0.8682292103767395,0.6990740895271301,0.9994792193174362,0.8694444596767426],"occluded":false},{"timestamp":9.36,"box":[0.9192708730697632,0.7249999642372131,0.9994792118668556,0.9055555164813995],"occluded":false},{"timestamp":9.620000000000001,"box":[0.96770840883255,0.7731481194496155,0.9994792453944683,0.9277777522802353],"occluded":false}],"mean_score":0.7104072570800781},{"noun_phrase":"massive boulder","object_id":3,"presentation_id":3,"observations":[{"timestamp":11.18,"box":[0.0,0.6851851940155029,0.1145833358168602,0.9731481373310089],"occluded":false},{"timestamp":11.44,"box":[0.0,0.6898148059844971,0.1770833432674408,0.9759259223937988],"occluded":false},{"timestamp":11.700000000000001,"box":[0.0,0.6722221970558167,0.22500000894069672,0.9740740358829498],"occluded":false},{"timestamp":11.96,"box":[0.0,0.6675925850868225,0.2692708373069763,0.9990740716457367],"occluded":false},{"timestamp":12.22,"box":[0.004166666883975267,0.654629647731781,0.9354167026467621,0.9990740716457367],"occluded":false},{"timestamp":12.48,"box":[0.08802083879709244,0.6361110806465149,0.33906251937150955,0.9638888537883759],"occluded":false},{"timestamp":12.74,"box":[0.11562500894069672,0.6203703880310059,0.35625001788139343,0.9407407641410828],"occluded":false}],"mean_score":0.7387754917144775},{"noun_phrase":"massive boulder","object_id":1,"presentation_id":4,"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":[0.22708334028720856,0.6287037134170532,0.46822918951511383,0.9648148119449615],"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":[0.31666669249534607,0.6296296119689941,0.4854166954755783,0.8675925731658936],"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":0.9139785170555115}],"dropped":{},"fates":[{"noun_phrase":"massive boulder","object_id":0,"frames":38,"mean_score":0.7104072570800781,"mean_area":0.02580193697679964,"outcome":"kept","presentation_id":2,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"massive boulder","object_id":1,"frames":44,"mean_score":0.9139785170555115,"mean_area":0.04375791539234503,"outcome":"kept","presentation_id":4,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"massive boulder","object_id":2,"frames":43,"mean_score":0.9569892287254333,"mean_area":0.1855980335099355,"outcome":"kept","presentation_id":1,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"massive boulder","object_id":3,"frames":7,"mean_score":0.7387754917144775,"mean_area":0.1029964458967543,"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":4,"as_video":true,"fps":1.0,"max_frames":30},"outputs":{"media":"/home/zzk/tmp/rmot_smoke_real_86260/renders/--cif7Nv6BI_chunk_012_scene_010_choices.mp4","media_bytes":5916470,"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","temperature":0.0,"as_video":true,"media_bytes":5916470,"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 massive boulder resting on top of the building with the red door\n2. the stone building with a red door underneath the giant boulder\n3. the black car driving down the street","questions":["the massive boulder resting on top of the building with the red door","the stone building with a red door underneath the giant boulder","the black car driving down 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\": 12,\n      \"confidence\": 1.0\n    },\n    {\n      \"description_index\": 2,\n      \"matching_ids\": [],\n      \"start_frame\": 1,\n      \"end_frame\": 14,\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":[1],"start_frame":1,"end_frame":12,"confidence":1.0},{"description_index":2,"matching_ids":[],"start_frame":1,"end_frame":14,"confidence":1.0},{"description_index":3,"matching_ids":[],"start_frame":1,"end_frame":14,"confidence":1.0}],"answers":[{"expression":"the massive boulder resting on top of the building with the red door","matching_presentation_ids":[1],"start_choice_frame":1,"end_choice_frame":12,"confidence":1.0,"self_reported_confidence":1.0},{"expression":"the stone building with a red door underneath the giant boulder","matching_presentation_ids":[],"start_choice_frame":1,"end_choice_frame":14,"confidence":1.0,"self_reported_confidence":1.0},{"expression":"the black car driving down the street","matching_presentation_ids":[],"start_choice_frame":1,"end_choice_frame":14,"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 massive boulder resting on top of the building with the red door","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],"boxes":[[0.3489583432674408,0.12037036567926407,0.7291666865348816,0.6268518343567848],[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],[0.25468751788139343,0.14814814925193787,0.7421875298023224,0.8009259402751923],[0.2562500238418579,0.14999999105930328,0.7427083849906921,0.7999999672174454],[0.2593750059604645,0.13148148357868195,0.7468750178813934,0.7805555611848831],[0.2588541805744171,0.0833333283662796,0.7463541924953461,0.7314814478158951],[0.2604166865348816,0.05000000074505806,0.7505208849906921,0.6972222216427326],[0.2588541805744171,0.031481482088565826,0.749479204416275,0.6675925627350807],[0.2614583373069763,0.031481482088565826,0.7463541924953461,0.6416666284203529],[0.27031251788139343,0.031481482088565826,0.7390625476837158,0.6194444075226784],[0.27656251192092896,0.03055555559694767,0.7250000238418579,0.6092592570930719],[0.2864583432674408,0.0416666641831398,0.7098958790302277,0.6064814701676369],[0.30677086114883423,0.069444440305233,0.705729216337204,0.5999999865889549],[0.3192708492279053,0.0962962955236435,0.6984375417232513,0.6018518656492233],[0.32864585518836975,0.11203703284263611,0.6921875476837158,0.5962962806224823],[0.3265625238418579,0.11481481045484543,0.6875000298023224,0.5962962731719017],[0.32343751192092896,0.11759258806705475,0.6848958730697632,0.5999999791383743],[0.3218750059604645,0.1240740716457367,0.6822916865348816,0.6027777791023254],[0.31979167461395264,0.13148148357868195,0.6802083551883698,0.6129629462957382],[0.31979167461395264,0.14907407760620117,0.6786458492279053,0.6268518567085266],[0.31979167461395264,0.16388888657093048,0.6786458492279053,0.6407407373189926],[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],[0.49531251192092896,0.2203703671693802,0.7890625298023224,0.5981481522321701],[0.526562511920929,0.2342592477798462,0.8161458671092987,0.6009259223937988],[0.5677083730697632,0.23981481790542603,0.861979216337204,0.6018518507480621],[0.6072916984558105,0.23888888955116272,0.9151042103767395,0.6027777791023254],[0.6453125476837158,0.2370370328426361,0.9739584028720856,0.6092592477798462],[0.6791666746139526,0.2361111044883728,0.999479204416275,0.6212962865829468],[0.7098958492279053,0.2361111044883728,0.999479204416275,0.6361111104488373],[0.7364583611488342,0.2361111044883728,0.999479204416275,0.652777761220932],[0.7812500596046448,0.23148147761821747,0.9994792342185974,0.6768518537282944],[0.8281250596046448,0.2481481432914734,0.9994792342185974,0.7064814567565918],[0.8822917342185974,0.27314814925193787,0.999479241669178,0.710185170173645]]}]}],"rejected":{"the black car driving down the street":"no_candidates_retrieved","the stone building with a red door underneath the giant boulder":"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_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 massive boulder resting on top of the building with the red door","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],"boxes":[[0.3489583432674408,0.12037036567926407,0.7291666865348816,0.6268518343567848],[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],[0.25468751788139343,0.14814814925193787,0.7421875298023224,0.8009259402751923],[0.2562500238418579,0.14999999105930328,0.7427083849906921,0.7999999672174454],[0.2593750059604645,0.13148148357868195,0.7468750178813934,0.7805555611848831],[0.2588541805744171,0.0833333283662796,0.7463541924953461,0.7314814478158951],[0.2604166865348816,0.05000000074505806,0.7505208849906921,0.6972222216427326],[0.2588541805744171,0.031481482088565826,0.749479204416275,0.6675925627350807],[0.2614583373069763,0.031481482088565826,0.7463541924953461,0.6416666284203529],[0.27031251788139343,0.031481482088565826,0.7390625476837158,0.6194444075226784],[0.27656251192092896,0.03055555559694767,0.7250000238418579,0.6092592570930719],[0.2864583432674408,0.0416666641831398,0.7098958790302277,0.6064814701676369],[0.30677086114883423,0.069444440305233,0.705729216337204,0.5999999865889549],[0.3192708492279053,0.0962962955236435,0.6984375417232513,0.6018518656492233],[0.32864585518836975,0.11203703284263611,0.6921875476837158,0.5962962806224823],[0.3265625238418579,0.11481481045484543,0.6875000298023224,0.5962962731719017],[0.32343751192092896,0.11759258806705475,0.6848958730697632,0.5999999791383743],[0.3218750059604645,0.1240740716457367,0.6822916865348816,0.6027777791023254],[0.31979167461395264,0.13148148357868195,0.6802083551883698,0.6129629462957382],[0.31979167461395264,0.14907407760620117,0.6786458492279053,0.6268518567085266],[0.31979167461395264,0.16388888657093048,0.6786458492279053,0.6407407373189926],[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],[0.49531251192092896,0.2203703671693802,0.7890625298023224,0.5981481522321701],[0.526562511920929,0.2342592477798462,0.8161458671092987,0.6009259223937988],[0.5677083730697632,0.23981481790542603,0.861979216337204,0.6018518507480621],[0.6072916984558105,0.23888888955116272,0.9151042103767395,0.6027777791023254],[0.6453125476837158,0.2370370328426361,0.9739584028720856,0.6092592477798462],[0.6791666746139526,0.2361111044883728,0.999479204416275,0.6212962865829468],[0.7098958492279053,0.2361111044883728,0.999479204416275,0.6361111104488373],[0.7364583611488342,0.2361111044883728,0.999479204416275,0.652777761220932],[0.7812500596046448,0.23148147761821747,0.9994792342185974,0.6768518537282944],[0.8281250596046448,0.2481481432914734,0.9994792342185974,0.7064814567565918],[0.8822917342185974,0.27314814925193787,0.999479241669178,0.710185170173645]]}]}],"rmot_expression_count":1,"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","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 bearded man walking down the street\",\n      \"noun_phrase\": \"bearded man\",\n      \"kind\": \"motion\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the silver car parked on the right side of the road\",\n      \"noun_phrase\": \"silver car\",\n      \"kind\": \"spatial\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"a person riding a bicycle down the street\",\n      \"noun_phrase\": \"cyclist\",\n      \"kind\": \"motion\",\n      \"expected_count\": 0,\n      \"is_negative\": true\n    }\n  ]\n}","raw_items":[{"expression":"the bearded man walking down the street","noun_phrase":"bearded man","kind":"motion","expected_count":1,"is_negative":false},{"expression":"the silver car parked on the right side of the road","noun_phrase":"silver car","kind":"spatial","expected_count":1,"is_negative":false},{"expression":"a person riding a bicycle down the street","noun_phrase":"cyclist","kind":"motion","expected_count":0,"is_negative":true}],"accepted":[{"expression":"the bearded man walking down the street","noun_phrase":"bearded man","kind":"motion","expected_count":1,"is_negative":false},{"expression":"the silver car parked on the right side of the road","noun_phrase":"silver car","kind":"spatial","expected_count":1,"is_negative":false},{"expression":"a person riding a bicycle down the street","noun_phrase":"cyclist","kind":"motion","expected_count":0,"is_negative":true}],"gated":[]}},"track":{"inputs":{"tracker":"Sam31MultiplexTracker","model_id":"/mnt/ceph3/ec/jacky/sam3.1","noun_phrases":["bearded man","cyclist","silver 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":"silver car","object_id":0,"observations":[{"timestamp":14.040000000000001,"box":[0.7854167222976685,0.5268518328666687,0.9083333984017372,0.6694444268941879],"occluded":false},{"timestamp":14.3,"box":[0.6739583611488342,0.5268518328666687,0.7807291969656944,0.6611110866069794],"occluded":false},{"timestamp":14.56,"box":[0.6229166984558105,0.5212962627410889,0.7250000387430191,0.6564814448356628],"occluded":false},{"timestamp":14.82,"box":[0.5973958373069763,0.5314814448356628,0.6994791775941849,0.6685184836387634],"occluded":false},{"timestamp":15.08,"box":[0.5807291865348816,0.5175926089286804,0.6828125268220901,0.661111131310463],"occluded":false},{"timestamp":15.34,"box":[0.5713542103767395,0.5212962627410889,0.674479216337204,0.6712962538003922],"occluded":false},{"timestamp":15.6,"box":[0.5473958849906921,0.5287036895751953,0.6526042222976685,0.6851851642131805],"occluded":false},{"timestamp":15.860000000000001,"box":[0.5557292103767395,0.5037037134170532,0.6625000461935997,0.6564814895391464],"occluded":false},{"timestamp":16.12,"box":[0.5604166984558105,0.5157407522201538,0.6734375357627869,0.6833333373069763],"occluded":false},{"timestamp":16.38,"box":[0.5760416984558105,0.5314814448356628,0.6963542029261589,0.700925886631012],"occluded":false},{"timestamp":16.64,"box":[0.5864583849906921,0.5305555462837219,0.7145833969116211,0.7129629552364349],"occluded":false},{"timestamp":16.9,"box":[0.659375011920929,0.5074074268341064,0.8119791895151138,0.7064815014600754],"occluded":false},{"timestamp":17.16,"box":[0.7442708611488342,0.4833333194255829,0.9354166984558105,0.7074073851108551],"occluded":false},{"timestamp":17.42,"box":[0.7713541984558105,0.5259259343147278,0.9932292103767395,0.7740740776062012],"occluded":false},{"timestamp":17.68,"box":[0.8187500238418579,0.5268518328666687,0.999479204416275,0.7620370090007782],"occluded":false},{"timestamp":17.94,"box":[0.8109375238418579,0.49537035822868347,0.999479204416275,0.7425925731658936],"occluded":false},{"timestamp":18.2,"box":[0.8114583492279053,0.49351850152015686,0.9994791895151138,0.7518518269062042],"occluded":false},{"timestamp":18.46,"box":[0.745312511920929,0.5685185194015503,0.999479204416275,0.8046296238899231],"occluded":false},{"timestamp":18.72,"box":[0.6505208611488342,0.5342592597007751,0.8875000327825546,0.8009259104728699],"occluded":false},{"timestamp":18.98,"box":[0.6140625476837158,0.5518518686294556,0.8531250655651093,0.8333333432674408],"occluded":false},{"timestamp":19.240000000000002,"box":[0.5895833373069763,0.5564814805984497,0.8364583551883698,0.8425925970077515],"occluded":false},{"timestamp":19.5,"box":[0.5770833492279053,0.555555522441864,0.8380208611488342,0.8537036776542664],"occluded":false},{"timestamp":19.76,"box":[0.5864583849906921,0.5351851582527161,0.8729167282581329,0.8472221791744232],"occluded":false},{"timestamp":20.02,"box":[0.6135417222976685,0.5342592597007751,0.947395920753479,0.8787036836147308],"occluded":false},{"timestamp":20.28,"box":[0.6817708611488342,0.5314814448356628,0.999479204416275,0.8685184717178345],"occluded":false},{"timestamp":20.54,"box":[0.7223958969116211,0.4879629611968994,0.9994792342185974,0.8731481432914734],"occluded":false},{"timestamp":20.8,"box":[0.7479166984558105,0.4564814865589142,0.999479204416275,0.8657407462596893],"occluded":false},{"timestamp":21.06,"box":[0.7848958969116211,0.4814814627170563,0.9994792342185974,0.9287036657333374],"occluded":false},{"timestamp":21.32,"box":[0.8218750357627869,0.4305555522441864,0.999479204416275,0.8907407224178314],"occluded":false},{"timestamp":21.580000000000002,"box":[0.8041667342185974,0.47592592239379883,0.9994792491197586,0.9574073851108551],"occluded":false},{"timestamp":21.84,"box":[0.7963542342185974,0.489814817905426,0.9994792491197586,0.9611110985279083],"occluded":false},{"timestamp":22.1,"box":[0.7546875476837158,0.46388888359069824,0.9994792342185974,0.9638888835906982],"occluded":false},{"timestamp":22.36,"box":[0.7598958611488342,0.5111110806465149,0.999479204416275,0.9712962508201599],"occluded":false},{"timestamp":22.62,"box":[0.737500011920929,0.5157407522201538,0.999479204416275,0.9731481373310089],"occluded":false},{"timestamp":22.88,"box":[0.7666667103767395,0.5222222208976746,0.9994792193174362,0.9740740656852722],"occluded":false},{"timestamp":23.14,"box":[0.8140625357627869,0.5842592716217041,0.9994792193174362,0.9731481671333313],"occluded":false},{"timestamp":23.400000000000002,"box":[0.8312500715255737,0.6231481432914734,0.9994792491197586,0.9722222089767456],"occluded":false},{"timestamp":23.66,"box":[0.8494791984558105,0.6962962746620178,0.999479204416275,0.9740740358829498],"occluded":false},{"timestamp":23.92,"box":[0.8812500238418579,0.6388888955116272,0.9994791969656944,0.9712963104248047],"occluded":false},{"timestamp":24.18,"box":[0.8927083611488342,0.7157407402992249,0.9994791969656944,0.9731481373310089],"occluded":false},{"timestamp":24.44,"box":[0.9609375596046448,0.7916666269302368,0.9994792267680168,0.9722221791744232],"occluded":false}],"scores":[0.8358974456787109,0.8358974456787109,0.8358974456787109,0.8358974456787109,0.8358974456787109,0.8358974456787109,0.8358974456787109,0.8358974456787109,0.8358974456787109,0.8358974456787109,0.8358974456787109,0.8358974456787109,0.8358974456787109,0.8358974456787109,0.8358974456787109,0.8358974456787109,0.8358974456787109,0.8358974456787109,0.8358974456787109,0.8358974456787109,0.8358974456787109,0.8358974456787109,0.8358974456787109,0.8358974456787109,0.8358974456787109,0.8358974456787109,0.8358974456787109,0.8358974456787109,0.8358974456787109,0.8358974456787109,0.8358974456787109,0.8358974456787109,0.8358974456787109,0.8358974456787109,0.8358974456787109,0.8358974456787109,0.8358974456787109,0.8358974456787109,0.8358974456787109,0.8358974456787109,0.8358974456787109]},{"noun_phrase":"silver car","object_id":1,"observations":[{"timestamp":16.64,"box":[0.5817708373069763,0.5259259343147278,0.6265625059604645,0.6342592686414719],"occluded":false},{"timestamp":16.9,"box":[0.6531250476837158,0.5018518567085266,0.6989583820104599,0.61296296864748],"occluded":false},{"timestamp":17.16,"box":[0.7343750596046448,0.48055553436279297,0.786458395421505,0.605555534362793],"occluded":false},{"timestamp":17.42,"box":[0.7572917342185974,0.5240740776062012,0.8109375685453415,0.6666666716337204],"occluded":false},{"timestamp":17.68,"box":[0.7979167103767395,0.5240740776062012,0.861979216337204,0.6712962985038757],"occluded":false},{"timestamp":17.94,"box":[0.7854167222976685,0.4990740716457367,0.8494792282581329,0.6537037044763565],"occluded":false},{"timestamp":18.2,"box":[0.7822917103767395,0.4990740716457367,0.8479167148470879,0.6564814746379852],"occluded":false},{"timestamp":18.46,"box":[0.7229167222976685,0.5583333373069763,0.7848958931863308,0.7037037014961243],"occluded":false},{"timestamp":18.72,"box":[0.6312500238418579,0.5314814448356628,0.684375025331974,0.6759258806705475],"occluded":false},{"timestamp":18.98,"box":[0.5916666984558105,0.5425925850868225,0.6479167006909847,0.686111107468605],"occluded":false},{"timestamp":19.240000000000002,"box":[0.5661458373069763,0.5425925850868225,0.6250000074505806,0.6898148059844971],"occluded":false},{"timestamp":19.5,"box":[0.550000011920929,0.5462962985038757,0.6114583499729633,0.6981481462717056],"occluded":false},{"timestamp":19.76,"box":[0.5562500357627869,0.5240740776062012,0.6229167059063911,0.6759259253740311],"occluded":false},{"timestamp":20.02,"box":[0.5765625238418579,0.5287036895751953,0.6510416939854622,0.6777777671813965],"occluded":false},{"timestamp":20.28,"box":[0.6328125596046448,0.5268518328666687,0.7223958969116211,0.6861110925674438],"occluded":false},{"timestamp":20.54,"box":[0.6661458611488342,0.4851851761341095,0.766145870089531,0.6472222059965134],"occluded":false},{"timestamp":20.8,"box":[0.6812500357627869,0.46481481194496155,0.7927083745598793,0.6370370239019394],"occluded":false},{"timestamp":21.06,"box":[0.7229167222976685,0.4981481432914734,0.8359375596046448,0.6509259194135666],"occluded":false},{"timestamp":21.32,"box":[0.7411458492279053,0.4564814865589142,0.8734375238418579,0.6194444447755814],"occluded":false},{"timestamp":21.580000000000002,"box":[0.7286458611488342,0.4981481432914734,0.8536458611488342,0.6666666567325592],"occluded":false},{"timestamp":21.84,"box":[0.714062511920929,0.49351850152015686,0.8562500178813934,0.662962943315506],"occluded":false},{"timestamp":22.1,"box":[0.6682292222976685,0.49166664481163025,0.8057292252779007,0.6601851582527161],"occluded":false},{"timestamp":22.36,"box":[0.6630208492279053,0.5287036895751953,0.8151041865348816,0.7074073851108551],"occluded":false},{"timestamp":22.62,"box":[0.6380208730697632,0.520370364189148,0.7953125536441803,0.7009259164333344],"occluded":false},{"timestamp":22.88,"box":[0.6484375596046448,0.529629647731781,0.8223959058523178,0.7194444537162781],"occluded":false},{"timestamp":23.14,"box":[0.6744791865348816,0.5249999761581421,0.8781250268220901,0.730555534362793],"occluded":false},{"timestamp":23.400000000000002,"box":[0.6739583611488342,0.5333333015441895,0.8859375417232513,0.7490740418434143],"occluded":false},{"timestamp":23.66,"box":[0.6708333492279053,0.5324074029922485,0.9000000208616257,0.7583333253860474],"occluded":false},{"timestamp":23.92,"box":[0.6781250238418579,0.46388888359069824,0.9255208671092987,0.7018518447875977],"occluded":false},{"timestamp":24.18,"box":[0.659375011920929,0.4981481432914734,0.917187511920929,0.7481481432914734],"occluded":false},{"timestamp":24.44,"box":[0.6744791865348816,0.5129629373550415,0.9781250357627869,0.8185184895992279],"occluded":false},{"timestamp":24.7,"box":[0.7005208730697632,0.5462962985038757,0.9994792342185974,0.8509259223937988],"occluded":false},{"timestamp":24.96,"box":[0.6802083849906921,0.4962962865829468,0.9994792342185974,0.8037036955356598],"occluded":false},{"timestamp":25.22,"box":[0.651562511920929,0.4888888895511627,0.999479204416275,0.8592592477798462],"occluded":false},{"timestamp":25.48,"box":[0.6041666865348816,0.5425925850868225,0.9630208611488342,0.8916666507720947],"occluded":false},{"timestamp":25.740000000000002,"box":[0.5651041865348816,0.5407407283782959,0.9218750298023224,0.8805555403232574],"occluded":false},{"timestamp":26.0,"box":[0.5807291865348816,0.5824074149131775,0.9921875298023224,0.9990740716457367],"occluded":false},{"timestamp":26.26,"box":[0.589062511920929,0.5842592716217041,0.999479204416275,0.9694444537162781],"occluded":false},{"timestamp":26.52,"box":[0.5895833373069763,0.5787037014961243,0.999479204416275,0.9694444239139557],"occluded":false},{"timestamp":26.78,"box":[0.6010416746139526,0.6194444298744202,0.999479204416275,0.9990740418434143],"occluded":false},{"timestamp":27.04,"box":[0.5958333611488342,0.6388888955116272,0.999479204416275,0.9990740716457367],"occluded":false},{"timestamp":27.3,"box":[0.5947917103767395,0.6009259223937988,0.9994792342185974,0.9731481373310089],"occluded":false},{"timestamp":27.560000000000002,"box":[0.5994791984558105,0.6277777552604675,0.9994792342185974,0.9787036776542664],"occluded":false},{"timestamp":27.82,"box":[0.5880208611488342,0.6287037134170532,0.999479204416275,0.9990740716457367],"occluded":false},{"timestamp":28.080000000000002,"box":[0.5901042222976685,0.6462962627410889,0.9994792342185974,0.9990740418434143],"occluded":false},{"timestamp":28.34,"box":[0.6041666865348816,0.6564814448356628,0.999479204416275,0.9722221791744232],"occluded":false},{"timestamp":28.6,"box":[0.6171875596046448,0.6435185074806213,0.9994792342185974,0.9990740418434143],"occluded":false},{"timestamp":28.86,"box":[0.620312511920929,0.6574074029922485,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":29.12,"box":[0.6270833611488342,0.6398147940635681,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":29.38,"box":[0.6307291984558105,0.6398147940635681,0.999479204416275,0.9740740358829498],"occluded":false},{"timestamp":29.64,"box":[0.6479166746139526,0.6407407522201538,0.999479204416275,0.972222238779068],"occluded":false},{"timestamp":29.900000000000002,"box":[0.8151041865348816,0.6740740537643433,0.9994791895151138,0.9731481373310089],"occluded":false}],"scores":[0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563,0.6559139490127563]}],"tracklets":[{"noun_phrase":"silver car","object_id":1,"presentation_id":1,"observations":[{"timestamp":16.64,"box":[0.5817708373069763,0.5259259343147278,0.6265625059604645,0.6342592686414719],"occluded":false},{"timestamp":16.9,"box":[0.6531250476837158,0.5018518567085266,0.6989583820104599,0.61296296864748],"occluded":false},{"timestamp":17.16,"box":[0.7343750596046448,0.48055553436279297,0.786458395421505,0.605555534362793],"occluded":false},{"timestamp":17.42,"box":[0.7572917342185974,0.5240740776062012,0.8109375685453415,0.6666666716337204],"occluded":false},{"timestamp":17.68,"box":[0.7979167103767395,0.5240740776062012,0.861979216337204,0.6712962985038757],"occluded":false},{"timestamp":17.94,"box":[0.7854167222976685,0.4990740716457367,0.8494792282581329,0.6537037044763565],"occluded":false},{"timestamp":18.2,"box":[0.7822917103767395,0.4990740716457367,0.8479167148470879,0.6564814746379852],"occluded":false},{"timestamp":18.46,"box":[0.7229167222976685,0.5583333373069763,0.7848958931863308,0.7037037014961243],"occluded":false},{"timestamp":18.72,"box":[0.6312500238418579,0.5314814448356628,0.684375025331974,0.6759258806705475],"occluded":false},{"timestamp":18.98,"box":[0.5916666984558105,0.5425925850868225,0.6479167006909847,0.686111107468605],"occluded":false},{"timestamp":19.240000000000002,"box":[0.5661458373069763,0.5425925850868225,0.6250000074505806,0.6898148059844971],"occluded":false},{"timestamp":19.5,"box":[0.550000011920929,0.5462962985038757,0.6114583499729633,0.6981481462717056],"occluded":false},{"timestamp":19.76,"box":[0.5562500357627869,0.5240740776062012,0.6229167059063911,0.6759259253740311],"occluded":false},{"timestamp":20.02,"box":[0.5765625238418579,0.5287036895751953,0.6510416939854622,0.6777777671813965],"occluded":false},{"timestamp":20.28,"box":[0.6328125596046448,0.5268518328666687,0.7223958969116211,0.6861110925674438],"occluded":false},{"timestamp":20.54,"box":[0.6661458611488342,0.4851851761341095,0.766145870089531,0.6472222059965134],"occluded":false},{"timestamp":20.8,"box":[0.6812500357627869,0.46481481194496155,0.7927083745598793,0.6370370239019394],"occluded":false},{"timestamp":21.06,"box":[0.7229167222976685,0.4981481432914734,0.8359375596046448,0.6509259194135666],"occluded":false},{"timestamp":21.32,"box":[0.7411458492279053,0.4564814865589142,0.8734375238418579,0.6194444447755814],"occluded":false},{"timestamp":21.580000000000002,"box":[0.7286458611488342,0.4981481432914734,0.8536458611488342,0.6666666567325592],"occluded":false},{"timestamp":21.84,"box":[0.714062511920929,0.49351850152015686,0.8562500178813934,0.662962943315506],"occluded":false},{"timestamp":22.1,"box":[0.6682292222976685,0.49166664481163025,0.8057292252779007,0.6601851582527161],"occluded":false},{"timestamp":22.36,"box":[0.6630208492279053,0.5287036895751953,0.8151041865348816,0.7074073851108551],"occluded":false},{"timestamp":22.62,"box":[0.6380208730697632,0.520370364189148,0.7953125536441803,0.7009259164333344],"occluded":false},{"timestamp":22.88,"box":[0.6484375596046448,0.529629647731781,0.8223959058523178,0.7194444537162781],"occluded":false},{"timestamp":23.14,"box":[0.6744791865348816,0.5249999761581421,0.8781250268220901,0.730555534362793],"occluded":false},{"timestamp":23.400000000000002,"box":[0.6739583611488342,0.5333333015441895,0.8859375417232513,0.7490740418434143],"occluded":false},{"timestamp":23.66,"box":[0.6708333492279053,0.5324074029922485,0.9000000208616257,0.7583333253860474],"occluded":false},{"timestamp":23.92,"box":[0.6781250238418579,0.46388888359069824,0.9255208671092987,0.7018518447875977],"occluded":false},{"timestamp":24.18,"box":[0.659375011920929,0.4981481432914734,0.917187511920929,0.7481481432914734],"occluded":false},{"timestamp":24.44,"box":[0.6744791865348816,0.5129629373550415,0.9781250357627869,0.8185184895992279],"occluded":false},{"timestamp":24.7,"box":[0.7005208730697632,0.5462962985038757,0.9994792342185974,0.8509259223937988],"occluded":false},{"timestamp":24.96,"box":[0.6802083849906921,0.4962962865829468,0.9994792342185974,0.8037036955356598],"occluded":false},{"timestamp":25.22,"box":[0.651562511920929,0.4888888895511627,0.999479204416275,0.8592592477798462],"occluded":false},{"timestamp":25.48,"box":[0.6041666865348816,0.5425925850868225,0.9630208611488342,0.8916666507720947],"occluded":false},{"timestamp":25.740000000000002,"box":[0.5651041865348816,0.5407407283782959,0.9218750298023224,0.8805555403232574],"occluded":false},{"timestamp":26.0,"box":[0.5807291865348816,0.5824074149131775,0.9921875298023224,0.9990740716457367],"occluded":false},{"timestamp":26.26,"box":[0.589062511920929,0.5842592716217041,0.999479204416275,0.9694444537162781],"occluded":false},{"timestamp":26.52,"box":[0.5895833373069763,0.5787037014961243,0.999479204416275,0.9694444239139557],"occluded":false},{"timestamp":26.78,"box":[0.6010416746139526,0.6194444298744202,0.999479204416275,0.9990740418434143],"occluded":false},{"timestamp":27.04,"box":[0.5958333611488342,0.6388888955116272,0.999479204416275,0.9990740716457367],"occluded":false},{"timestamp":27.3,"box":[0.5947917103767395,0.6009259223937988,0.9994792342185974,0.9731481373310089],"occluded":false},{"timestamp":27.560000000000002,"box":[0.5994791984558105,0.6277777552604675,0.9994792342185974,0.9787036776542664],"occluded":false},{"timestamp":27.82,"box":[0.5880208611488342,0.6287037134170532,0.999479204416275,0.9990740716457367],"occluded":false},{"timestamp":28.080000000000002,"box":[0.5901042222976685,0.6462962627410889,0.9994792342185974,0.9990740418434143],"occluded":false},{"timestamp":28.34,"box":[0.6041666865348816,0.6564814448356628,0.999479204416275,0.9722221791744232],"occluded":false},{"timestamp":28.6,"box":[0.6171875596046448,0.6435185074806213,0.9994792342185974,0.9990740418434143],"occluded":false},{"timestamp":28.86,"box":[0.620312511920929,0.6574074029922485,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":29.12,"box":[0.6270833611488342,0.6398147940635681,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":29.38,"box":[0.6307291984558105,0.6398147940635681,0.999479204416275,0.9740740358829498],"occluded":false},{"timestamp":29.64,"box":[0.6479166746139526,0.6407407522201538,0.999479204416275,0.972222238779068],"occluded":false},{"timestamp":29.900000000000002,"box":[0.8151041865348816,0.6740740537643433,0.9994791895151138,0.9731481373310089],"occluded":false}],"mean_score":0.6559139490127563},{"noun_phrase":"bearded man","object_id":0,"presentation_id":2,"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},{"noun_phrase":"silver car","object_id":0,"presentation_id":3,"observations":[{"timestamp":14.040000000000001,"box":[0.7854167222976685,0.5268518328666687,0.9083333984017372,0.6694444268941879],"occluded":false},{"timestamp":14.3,"box":[0.6739583611488342,0.5268518328666687,0.7807291969656944,0.6611110866069794],"occluded":false},{"timestamp":14.56,"box":[0.6229166984558105,0.5212962627410889,0.7250000387430191,0.6564814448356628],"occluded":false},{"timestamp":14.82,"box":[0.5973958373069763,0.5314814448356628,0.6994791775941849,0.6685184836387634],"occluded":false},{"timestamp":15.08,"box":[0.5807291865348816,0.5175926089286804,0.6828125268220901,0.661111131310463],"occluded":false},{"timestamp":15.34,"box":[0.5713542103767395,0.5212962627410889,0.674479216337204,0.6712962538003922],"occluded":false},{"timestamp":15.6,"box":[0.5473958849906921,0.5287036895751953,0.6526042222976685,0.6851851642131805],"occluded":false},{"timestamp":15.860000000000001,"box":[0.5557292103767395,0.5037037134170532,0.6625000461935997,0.6564814895391464],"occluded":false},{"timestamp":16.12,"box":[0.5604166984558105,0.5157407522201538,0.6734375357627869,0.6833333373069763],"occluded":false},{"timestamp":16.38,"box":[0.5760416984558105,0.5314814448356628,0.6963542029261589,0.700925886631012],"occluded":false},{"timestamp":16.64,"box":[0.5864583849906921,0.5305555462837219,0.7145833969116211,0.7129629552364349],"occluded":false},{"timestamp":16.9,"box":[0.659375011920929,0.5074074268341064,0.8119791895151138,0.7064815014600754],"occluded":false},{"timestamp":17.16,"box":[0.7442708611488342,0.4833333194255829,0.9354166984558105,0.7074073851108551],"occluded":false},{"timestamp":17.42,"box":[0.7713541984558105,0.5259259343147278,0.9932292103767395,0.7740740776062012],"occluded":false},{"timestamp":17.68,"box":[0.8187500238418579,0.5268518328666687,0.999479204416275,0.7620370090007782],"occluded":false},{"timestamp":17.94,"box":[0.8109375238418579,0.49537035822868347,0.999479204416275,0.7425925731658936],"occluded":false},{"timestamp":18.2,"box":[0.8114583492279053,0.49351850152015686,0.9994791895151138,0.7518518269062042],"occluded":false},{"timestamp":18.46,"box":[0.745312511920929,0.5685185194015503,0.999479204416275,0.8046296238899231],"occluded":false},{"timestamp":18.72,"box":[0.6505208611488342,0.5342592597007751,0.8875000327825546,0.8009259104728699],"occluded":false},{"timestamp":18.98,"box":[0.6140625476837158,0.5518518686294556,0.8531250655651093,0.8333333432674408],"occluded":false},{"timestamp":19.240000000000002,"box":[0.5895833373069763,0.5564814805984497,0.8364583551883698,0.8425925970077515],"occluded":false},{"timestamp":19.5,"box":[0.5770833492279053,0.555555522441864,0.8380208611488342,0.8537036776542664],"occluded":false},{"timestamp":19.76,"box":[0.5864583849906921,0.5351851582527161,0.8729167282581329,0.8472221791744232],"occluded":false},{"timestamp":20.02,"box":[0.6135417222976685,0.5342592597007751,0.947395920753479,0.8787036836147308],"occluded":false},{"timestamp":20.28,"box":[0.6817708611488342,0.5314814448356628,0.999479204416275,0.8685184717178345],"occluded":false},{"timestamp":20.54,"box":[0.7223958969116211,0.4879629611968994,0.9994792342185974,0.8731481432914734],"occluded":false},{"timestamp":20.8,"box":[0.7479166984558105,0.4564814865589142,0.999479204416275,0.8657407462596893],"occluded":false},{"timestamp":21.06,"box":[0.7848958969116211,0.4814814627170563,0.9994792342185974,0.9287036657333374],"occluded":false},{"timestamp":21.32,"box":[0.8218750357627869,0.4305555522441864,0.999479204416275,0.8907407224178314],"occluded":false},{"timestamp":21.580000000000002,"box":[0.8041667342185974,0.47592592239379883,0.9994792491197586,0.9574073851108551],"occluded":false},{"timestamp":21.84,"box":[0.7963542342185974,0.489814817905426,0.9994792491197586,0.9611110985279083],"occluded":false},{"timestamp":22.1,"box":[0.7546875476837158,0.46388888359069824,0.9994792342185974,0.9638888835906982],"occluded":false},{"timestamp":22.36,"box":[0.7598958611488342,0.5111110806465149,0.999479204416275,0.9712962508201599],"occluded":false},{"timestamp":22.62,"box":[0.737500011920929,0.5157407522201538,0.999479204416275,0.9731481373310089],"occluded":false},{"timestamp":22.88,"box":[0.7666667103767395,0.5222222208976746,0.9994792193174362,0.9740740656852722],"occluded":false},{"timestamp":23.14,"box":[0.8140625357627869,0.5842592716217041,0.9994792193174362,0.9731481671333313],"occluded":false},{"timestamp":23.400000000000002,"box":[0.8312500715255737,0.6231481432914734,0.9994792491197586,0.9722222089767456],"occluded":false},{"timestamp":23.66,"box":[0.8494791984558105,0.6962962746620178,0.999479204416275,0.9740740358829498],"occluded":false},{"timestamp":23.92,"box":[0.8812500238418579,0.6388888955116272,0.9994791969656944,0.9712963104248047],"occluded":false},{"timestamp":24.18,"box":[0.8927083611488342,0.7157407402992249,0.9994791969656944,0.9731481373310089],"occluded":false},{"timestamp":24.44,"box":[0.9609375596046448,0.7916666269302368,0.9994792267680168,0.9722221791744232],"occluded":false}],"mean_score":0.8358974456787109}],"dropped":{},"fates":[{"noun_phrase":"bearded man","object_id":0,"frames":11,"mean_score":0.7037037014961243,"mean_area":0.1083792835631638,"outcome":"kept","presentation_id":2,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"silver car","object_id":0,"frames":41,"mean_score":0.8358974456787109,"mean_area":0.05811362764297803,"outcome":"kept","presentation_id":3,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"silver car","object_id":1,"frames":52,"mean_score":0.6559139490127563,"mean_area":0.06646549483032291,"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":3,"as_video":true,"fps":1.0,"max_frames":30},"outputs":{"media":"/home/zzk/tmp/rmot_smoke_real_86260/renders/--cif7Nv6BI_chunk_012_scene_017_choices.mp4","media_bytes":7906594,"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","temperature":0.0,"as_video":true,"media_bytes":7906594,"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 bearded man walking down the street\n2. the silver car parked on the right side of the road\n3. a person riding a bicycle down the street","questions":["the bearded man walking down the street","the silver car parked on the right side of the road","a person riding a bicycle down the street"]},"outputs":{"raw_response":"{\n  \"answers\": [\n    {\n      \"description_index\": 1,\n      \"matching_ids\": [\n        2\n      ],\n      \"start_frame\": 2,\n      \"end_frame\": 4,\n      \"confidence\": 0.9\n    },\n    {\n      \"description_index\": 2,\n      \"matching_ids\": [\n        1,\n        3\n      ],\n      \"start_frame\": 12,\n      \"end_frame\": 22,\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":[2],"start_frame":2,"end_frame":4,"confidence":0.9},{"description_index":2,"matching_ids":[1,3],"start_frame":12,"end_frame":22,"confidence":0.95},{"description_index":3,"matching_ids":[],"start_frame":1,"end_frame":30,"confidence":1.0}],"answers":[{"expression":"the bearded man walking down the street","matching_presentation_ids":[2],"start_choice_frame":2,"end_choice_frame":4,"confidence":0.9,"self_reported_confidence":0.9},{"expression":"the silver car parked on the right side of the road","matching_presentation_ids":[1,3],"start_choice_frame":12,"end_choice_frame":22,"confidence":0.95,"self_reported_confidence":0.95},{"expression":"a person riding a bicycle down the 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 bearded man walking down the street","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]]}]}],"rejected":{"a person riding a bicycle down the street":"no_candidates_retrieved","the silver car parked on the right side of the road":"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_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 bearded man walking down the street","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]]}]}],"rmot_expression_count":1,"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","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 on the table\",\n      \"noun_phrase\": \"white teapot\",\n      \"kind\": \"spatial\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the stone building visible through the window\",\n      \"noun_phrase\": \"stone building\",\n      \"kind\": \"spatial\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the red cup on the table\",\n      \"noun_phrase\": \"red cup\",\n      \"kind\": \"appearance\",\n      \"expected_count\": 0,\n      \"is_negative\": true\n    }\n  ]\n}","raw_items":[{"expression":"the white teapot on the table","noun_phrase":"white teapot","kind":"spatial","expected_count":1,"is_negative":false},{"expression":"the stone building visible through the window","noun_phrase":"stone building","kind":"spatial","expected_count":1,"is_negative":false},{"expression":"the red cup on the table","noun_phrase":"red cup","kind":"appearance","expected_count":0,"is_negative":true}],"accepted":[{"expression":"the white teapot on the table","noun_phrase":"white teapot","kind":"spatial","expected_count":1,"is_negative":false},{"expression":"the stone building visible through the window","noun_phrase":"stone building","kind":"spatial","expected_count":1,"is_negative":false},{"expression":"the red cup on the table","noun_phrase":"red cup","kind":"appearance","expected_count":0,"is_negative":true}],"gated":[]}},"track":{"inputs":{"tracker":"Sam31MultiplexTracker","model_id":"/mnt/ceph3/ec/jacky/sam3.1","noun_phrases":["red cup","stone building","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":"stone building","object_id":0,"observations":[{"timestamp":5.46,"box":[0.49531251192092896,0.031481482088565826,0.9994792342185974,0.36574072390794754],"occluded":false},{"timestamp":5.72,"box":[0.2875000238418579,0.031481482088565826,0.84270840883255,0.49166665226221085],"occluded":false},{"timestamp":5.98,"box":[0.08958333730697632,0.03055555559694767,0.6838542222976685,0.6601851675659418],"occluded":false},{"timestamp":6.24,"box":[0.0,0.031481482088565826,0.604687511920929,0.7685184851288795],"occluded":false},{"timestamp":6.5,"box":[0.0,0.03611110895872116,0.5515625476837158,0.8527777716517448],"occluded":false},{"timestamp":6.76,"box":[0.0,0.08055555075407028,0.5208333730697632,0.8981481119990349],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.0,0.1398148089647293,0.5015625357627869,0.9287036806344986],"occluded":false},{"timestamp":7.28,"box":[0.0,0.1537037044763565,0.4854166805744171,0.9351851493120193],"occluded":false},{"timestamp":7.54,"box":[0.0,0.16111110150814056,0.4697916805744171,0.9361110776662827],"occluded":false},{"timestamp":7.8,"box":[0.0,0.16111110150814056,0.43385419249534607,0.9148148149251938],"occluded":false},{"timestamp":8.06,"box":[0.07708333432674408,0.14907407760620117,0.3635416775941849,0.8916666507720947],"occluded":false},{"timestamp":8.32,"box":[0.0,0.11481481045484543,0.24635417759418488,0.9138888642191887],"occluded":false},{"timestamp":8.58,"box":[0.0,0.2916666567325592,0.0989583358168602,0.8601851761341095],"occluded":false}],"scores":[0.7968254089355469,0.7968254089355469,0.7968254089355469,0.7968254089355469,0.7968254089355469,0.7968254089355469,0.7968254089355469,0.7968254089355469,0.7968254089355469,0.7968254089355469,0.7968254089355469,0.7968254089355469,0.7968254089355469]},{"noun_phrase":"stone building","object_id":1,"observations":[{"timestamp":9.88,"box":[0.7895833849906921,0.37962961196899414,0.9583333879709244,0.8564814627170563],"occluded":false},{"timestamp":10.14,"box":[0.7552083730697632,0.37870368361473083,0.9994792193174362,0.8499999642372131],"occluded":false},{"timestamp":10.4,"box":[0.7270833849906921,0.37870368361473083,0.9994792342185974,0.8472221791744232],"occluded":false}],"scores":[0.6952381134033203,0.6952381134033203,0.6952381134033203]},{"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":"white teapot","object_id":0,"presentation_id":1,"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":"stone building","object_id":1,"presentation_id":2,"observations":[{"timestamp":9.88,"box":[0.7895833849906921,0.37962961196899414,0.9583333879709244,0.8564814627170563],"occluded":false},{"timestamp":10.14,"box":[0.7552083730697632,0.37870368361473083,0.9994792193174362,0.8499999642372131],"occluded":false},{"timestamp":10.4,"box":[0.7270833849906921,0.37870368361473083,0.9994792342185974,0.8472221791744232],"occluded":false}],"mean_score":0.6952381134033203},{"noun_phrase":"stone building","object_id":0,"presentation_id":3,"observations":[{"timestamp":5.46,"box":[0.49531251192092896,0.031481482088565826,0.9994792342185974,0.36574072390794754],"occluded":false},{"timestamp":5.72,"box":[0.2875000238418579,0.031481482088565826,0.84270840883255,0.49166665226221085],"occluded":false},{"timestamp":5.98,"box":[0.08958333730697632,0.03055555559694767,0.6838542222976685,0.6601851675659418],"occluded":false},{"timestamp":6.24,"box":[0.0,0.031481482088565826,0.604687511920929,0.7685184851288795],"occluded":false},{"timestamp":6.5,"box":[0.0,0.03611110895872116,0.5515625476837158,0.8527777716517448],"occluded":false},{"timestamp":6.76,"box":[0.0,0.08055555075407028,0.5208333730697632,0.8981481119990349],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.0,0.1398148089647293,0.5015625357627869,0.9287036806344986],"occluded":false},{"timestamp":7.28,"box":[0.0,0.1537037044763565,0.4854166805744171,0.9351851493120193],"occluded":false},{"timestamp":7.54,"box":[0.0,0.16111110150814056,0.4697916805744171,0.9361110776662827],"occluded":false},{"timestamp":7.8,"box":[0.0,0.16111110150814056,0.43385419249534607,0.9148148149251938],"occluded":false},{"timestamp":8.06,"box":[0.07708333432674408,0.14907407760620117,0.3635416775941849,0.8916666507720947],"occluded":false},{"timestamp":8.32,"box":[0.0,0.11481481045484543,0.24635417759418488,0.9138888642191887],"occluded":false},{"timestamp":8.58,"box":[0.0,0.2916666567325592,0.0989583358168602,0.8601851761341095],"occluded":false}],"mean_score":0.7968254089355469}],"dropped":{},"fates":[{"noun_phrase":"stone building","object_id":0,"frames":13,"mean_score":0.7968254089355469,"mean_area":0.3116988393429269,"outcome":"kept","presentation_id":3,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"stone building","object_id":1,"frames":3,"mean_score":0.6952381134033203,"mean_area":0.10773839533762075,"outcome":"kept","presentation_id":2,"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":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":3,"as_video":true,"fps":1.0,"max_frames":30},"outputs":{"media":"/home/zzk/tmp/rmot_smoke_real_86260/renders/--cif7Nv6BI_chunk_000_scene_042_choices.mp4","media_bytes":2996535,"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","temperature":0.0,"as_video":true,"media_bytes":2996535,"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 on the table\n2. the stone building visible through the window\n3. the red cup on the table","questions":["the white teapot on the table","the stone building visible through the window","the red cup on the table"]},"outputs":{"raw_response":"{\n  \"answers\": [\n    {\n      \"description_index\": 1,\n      \"matching_ids\": [\n        1\n      ],\n      \"start_frame\": 1,\n      \"end_frame\": 5,\n      \"confidence\": 1.0\n    },\n    {\n      \"description_index\": 2,\n      \"matching_ids\": [\n        3\n      ],\n      \"start_frame\": 7,\n      \"end_frame\": 9,\n      \"confidence\": 0.9\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":[1],"start_frame":1,"end_frame":5,"confidence":1.0},{"description_index":2,"matching_ids":[3],"start_frame":7,"end_frame":9,"confidence":0.9},{"description_index":3,"matching_ids":[],"start_frame":1,"end_frame":11,"confidence":1.0}],"answers":[{"expression":"the white teapot on the table","matching_presentation_ids":[1],"start_choice_frame":1,"end_choice_frame":5,"confidence":1.0,"self_reported_confidence":1.0},{"expression":"the stone building visible through the window","matching_presentation_ids":[3],"start_choice_frame":7,"end_choice_frame":9,"confidence":0.9,"self_reported_confidence":0.9},{"expression":"the red cup on the table","matching_presentation_ids":[],"start_choice_frame":1,"end_choice_frame":11,"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 white teapot on the table","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],"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 stone building visible through the window","kind":"spatial","is_negative":false,"tracks":[{"timestamps":[6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32],"boxes":[[0.0,0.031481482088565826,0.604687511920929,0.7685184851288795],[0.0,0.03611110895872116,0.5515625476837158,0.8527777716517448],[0.0,0.08055555075407028,0.5208333730697632,0.8981481119990349],[0.0,0.1398148089647293,0.5015625357627869,0.9287036806344986],[0.0,0.1537037044763565,0.4854166805744171,0.9351851493120193],[0.0,0.16111110150814056,0.4697916805744171,0.9361110776662827],[0.0,0.16111110150814056,0.43385419249534607,0.9148148149251938],[0.07708333432674408,0.14907407760620117,0.3635416775941849,0.8916666507720947],[0.0,0.11481481045484543,0.24635417759418488,0.9138888642191887]]}]}],"rejected":{"the red cup on the table":"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 on the table","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],"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 stone building visible through the window","kind":"spatial","is_negative":false,"tracks":[{"timestamps":[6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32],"boxes":[[0.0,0.031481482088565826,0.604687511920929,0.7685184851288795],[0.0,0.03611110895872116,0.5515625476837158,0.8527777716517448],[0.0,0.08055555075407028,0.5208333730697632,0.8981481119990349],[0.0,0.1398148089647293,0.5015625357627869,0.9287036806344986],[0.0,0.1537037044763565,0.4854166805744171,0.9351851493120193],[0.0,0.16111110150814056,0.4697916805744171,0.9361110776662827],[0.0,0.16111110150814056,0.43385419249534607,0.9148148149251938],[0.07708333432674408,0.14907407760620117,0.3635416775941849,0.8916666507720947],[0.0,0.11481481045484543,0.24635417759418488,0.9138888642191887]]}]}],"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","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 balcony with the green metal railing on the upper floor\",\n      \"noun_phrase\": \"balcony\",\n      \"kind\": \"appearance\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the white van parked on the street\",\n      \"noun_phrase\": \"white van\",\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 balcony with the green metal railing on the upper floor","noun_phrase":"balcony","kind":"appearance","expected_count":1,"is_negative":false},{"expression":"the white van parked on the street","noun_phrase":"white van","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 balcony with the green metal railing on the upper floor","noun_phrase":"balcony","kind":"appearance","expected_count":1,"is_negative":false},{"expression":"the white van parked on the street","noun_phrase":"white van","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":["balcony","red car","white van"],"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":"balcony","object_id":0,"observations":[{"timestamp":0.0,"box":[0.6093750596046448,0.125,0.9994792342185974,0.6083333194255829],"occluded":false},{"timestamp":0.26,"box":[0.596875011920929,0.12777777016162872,0.999479204416275,0.5999999791383743],"occluded":false},{"timestamp":0.52,"box":[0.5765625238418579,0.1509259194135666,0.999479204416275,0.5879629403352737],"occluded":false},{"timestamp":0.78,"box":[0.557812511920929,0.1814814805984497,0.999479204416275,0.5768518447875977],"occluded":false},{"timestamp":1.04,"box":[0.5244792103767395,0.22962962090969086,0.9994792342185974,0.5796296149492264],"occluded":false},{"timestamp":1.3,"box":[0.4828125238418579,0.21111111342906952,0.9994792342185974,0.5898147970438004],"occluded":false},{"timestamp":1.56,"box":[0.4281250238418579,0.14722222089767456,0.9171875417232513,0.567592591047287],"occluded":false},{"timestamp":1.82,"box":[0.37187501788139343,0.07129629701375961,0.8317708671092987,0.5259259268641472],"occluded":false},{"timestamp":2.08,"box":[0.31041666865348816,0.03055555559694767,0.7604166865348816,0.4648148212581873],"occluded":false},{"timestamp":2.34,"box":[0.2510416805744171,0.029629629105329514,0.7067708671092987,0.4018518440425396],"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":[0.18281251192092896,0.029629629105329514,0.6151041984558105,0.24629629775881767],"occluded":false},{"timestamp":3.38,"box":[0.1953125149011612,0.03055555559694767,0.5843750387430191,0.23703702725470066],"occluded":false},{"timestamp":3.64,"box":[0.2317708432674408,0.03055555559694767,0.5625000298023224,0.2546296212822199],"occluded":false},{"timestamp":3.9,"box":[0.21927084028720856,0.03055555559694767,0.5500000268220901,0.2703703735023737],"occluded":false},{"timestamp":4.16,"box":[0.1536458432674408,0.03055555559694767,0.5500000417232513,0.2824074123054743],"occluded":false},{"timestamp":4.42,"box":[0.09531250596046448,0.029629629105329514,0.5838541984558105,0.32870371267199516],"occluded":false},{"timestamp":4.68,"box":[0.19583334028720856,0.031481482088565826,0.9911458641290665,0.7481481209397316],"occluded":false},{"timestamp":4.94,"box":[0.30885419249534607,0.3027777671813965,0.7734375596046448,0.7527777552604675],"occluded":false},{"timestamp":5.2,"box":[0.36145836114883423,0.5064814686775208,0.8692708611488342,0.930555522441864],"occluded":false},{"timestamp":5.46,"box":[0.3854166865348816,0.5361111164093018,0.9770833849906921,0.9703703820705414],"occluded":false},{"timestamp":5.72,"box":[0.38697919249534607,0.5740740895271301,0.999479204416275,0.9712963104248047],"occluded":false},{"timestamp":5.98,"box":[0.37447917461395264,0.595370352268219,0.9994792342185974,0.9722222089767456],"occluded":false},{"timestamp":6.24,"box":[0.3760416805744171,0.5972222089767456,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":6.5,"box":[0.3697916865348816,0.5907407402992249,0.9994792342185974,0.9712962806224823],"occluded":false},{"timestamp":6.76,"box":[0.36250001192092896,0.5277777910232544,0.9994792342185974,0.9712963104248047],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.3125000298023224,0.5120370388031006,0.999479204416275,0.9712962806224823],"occluded":false},{"timestamp":7.28,"box":[0.20364584028720856,0.4870370328426361,0.8578125387430191,0.9722222089767456],"occluded":false},{"timestamp":7.54,"box":[0.11093750596046448,0.402777761220932,0.7557292282581329,0.9722221791744232],"occluded":false},{"timestamp":7.8,"box":[0.05260416865348816,0.32499998807907104,0.6937500536441803,0.970370352268219],"occluded":false},{"timestamp":8.06,"box":[0.04062500223517418,0.27129629254341125,0.6770833395421505,0.9703703820705414],"occluded":false},{"timestamp":8.32,"box":[0.08906250447034836,0.2768518328666687,0.7046875283122063,0.9694443941116333],"occluded":false},{"timestamp":8.58,"box":[0.1875000149011612,0.38333332538604736,0.8187500387430191,0.970370352268219],"occluded":false},{"timestamp":8.84,"box":[0.31302085518836975,0.41111111640930176,0.9666667282581329,0.970370352268219],"occluded":false},{"timestamp":9.1,"box":[0.4203125238418579,0.29629629850387573,0.9994792342185974,0.8185185194015503],"occluded":false},{"timestamp":9.36,"box":[0.6244791746139526,0.029629629105329514,0.999479204416275,0.46388889476656914],"occluded":false}],"scores":[0.8947368264198303,0.8947368264198303,0.8947368264198303,0.8947368264198303,0.8947368264198303,0.8947368264198303,0.8947368264198303,0.8947368264198303,0.8947368264198303,0.8947368264198303,0.8947368264198303,0.8947368264198303,0.8947368264198303,0.8947368264198303,0.8947368264198303,0.8947368264198303,0.8947368264198303,0.8947368264198303,0.8947368264198303,0.8947368264198303,0.8947368264198303,0.8947368264198303,0.8947368264198303,0.8947368264198303,0.8947368264198303,0.8947368264198303,0.8947368264198303,0.8947368264198303,0.8947368264198303,0.8947368264198303,0.8947368264198303,0.8947368264198303,0.8947368264198303,0.8947368264198303,0.8947368264198303,0.8947368264198303,0.8947368264198303]},{"noun_phrase":"balcony","object_id":1,"observations":[{"timestamp":2.86,"box":[0.9093750715255737,0.4888888895511627,0.999479241669178,0.5814814791083336],"occluded":false},{"timestamp":3.12,"box":[0.8656250238418579,0.4583333134651184,0.9468750283122063,0.5462962761521339],"occluded":false},{"timestamp":3.38,"box":[0.8145833611488342,0.46481481194496155,0.8854167014360428,0.5481481403112411],"occluded":false},{"timestamp":3.64,"box":[0.7776042222976685,0.510185182094574,0.8427083939313889,0.5851851776242256],"occluded":false},{"timestamp":3.9,"box":[0.7572917342185974,0.5361111164093018,0.8182292357087135,0.61296296864748],"occluded":false},{"timestamp":4.16,"box":[0.7588542103767395,0.5462962985038757,0.8218750506639481,0.6259259283542633],"occluded":false},{"timestamp":4.42,"box":[0.8109375238418579,0.5712962746620178,0.8838541954755783,0.6574073806405067],"occluded":false}],"scores":[0.7299578189849854,0.7299578189849854,0.7299578189849854,0.7299578189849854,0.7299578189849854,0.7299578189849854,0.7299578189849854]},{"noun_phrase":"white van","object_id":0,"observations":[{"timestamp":10.4,"box":[0.2395833432674408,0.5564814805984497,0.3333333507180214,0.706481471657753],"occluded":false},{"timestamp":10.66,"box":[0.28072917461395264,0.539814829826355,0.369791679084301,0.6777777969837189],"occluded":false},{"timestamp":10.92,"box":[0.30156251788139343,0.5027777552604675,0.3906250223517418,0.6379629373550415],"occluded":false},{"timestamp":11.18,"box":[0.31822916865348816,0.5166666507720947,0.4078125059604645,0.6537036895751953],"occluded":false},{"timestamp":11.44,"box":[0.33020836114883423,0.5259259343147278,0.4192708656191826,0.6638889014720917],"occluded":false}],"scores":[0.6952381134033203,0.6952381134033203,0.6952381134033203,0.6952381134033203,0.6952381134033203]}],"tracklets":[{"noun_phrase":"balcony","object_id":1,"presentation_id":1,"observations":[{"timestamp":2.86,"box":[0.9093750715255737,0.4888888895511627,0.999479241669178,0.5814814791083336],"occluded":false},{"timestamp":3.12,"box":[0.8656250238418579,0.4583333134651184,0.9468750283122063,0.5462962761521339],"occluded":false},{"timestamp":3.38,"box":[0.8145833611488342,0.46481481194496155,0.8854167014360428,0.5481481403112411],"occluded":false},{"timestamp":3.64,"box":[0.7776042222976685,0.510185182094574,0.8427083939313889,0.5851851776242256],"occluded":false},{"timestamp":3.9,"box":[0.7572917342185974,0.5361111164093018,0.8182292357087135,0.61296296864748],"occluded":false},{"timestamp":4.16,"box":[0.7588542103767395,0.5462962985038757,0.8218750506639481,0.6259259283542633],"occluded":false},{"timestamp":4.42,"box":[0.8109375238418579,0.5712962746620178,0.8838541954755783,0.6574073806405067],"occluded":false}],"mean_score":0.7299578189849854},{"noun_phrase":"white van","object_id":0,"presentation_id":2,"observations":[{"timestamp":10.4,"box":[0.2395833432674408,0.5564814805984497,0.3333333507180214,0.706481471657753],"occluded":false},{"timestamp":10.66,"box":[0.28072917461395264,0.539814829826355,0.369791679084301,0.6777777969837189],"occluded":false},{"timestamp":10.92,"box":[0.30156251788139343,0.5027777552604675,0.3906250223517418,0.6379629373550415],"occluded":false},{"timestamp":11.18,"box":[0.31822916865348816,0.5166666507720947,0.4078125059604645,0.6537036895751953],"occluded":false},{"timestamp":11.44,"box":[0.33020836114883423,0.5259259343147278,0.4192708656191826,0.6638889014720917],"occluded":false}],"mean_score":0.6952381134033203},{"noun_phrase":"balcony","object_id":0,"presentation_id":3,"observations":[{"timestamp":0.0,"box":[0.6093750596046448,0.125,0.9994792342185974,0.6083333194255829],"occluded":false},{"timestamp":0.26,"box":[0.596875011920929,0.12777777016162872,0.999479204416275,0.5999999791383743],"occluded":false},{"timestamp":0.52,"box":[0.5765625238418579,0.1509259194135666,0.999479204416275,0.5879629403352737],"occluded":false},{"timestamp":0.78,"box":[0.557812511920929,0.1814814805984497,0.999479204416275,0.5768518447875977],"occluded":false},{"timestamp":1.04,"box":[0.5244792103767395,0.22962962090969086,0.9994792342185974,0.5796296149492264],"occluded":false},{"timestamp":1.3,"box":[0.4828125238418579,0.21111111342906952,0.9994792342185974,0.5898147970438004],"occluded":false},{"timestamp":1.56,"box":[0.4281250238418579,0.14722222089767456,0.9171875417232513,0.567592591047287],"occluded":false},{"timestamp":1.82,"box":[0.37187501788139343,0.07129629701375961,0.8317708671092987,0.5259259268641472],"occluded":false},{"timestamp":2.08,"box":[0.31041666865348816,0.03055555559694767,0.7604166865348816,0.4648148212581873],"occluded":false},{"timestamp":2.34,"box":[0.2510416805744171,0.029629629105329514,0.7067708671092987,0.4018518440425396],"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":[0.18281251192092896,0.029629629105329514,0.6151041984558105,0.24629629775881767],"occluded":false},{"timestamp":3.38,"box":[0.1953125149011612,0.03055555559694767,0.5843750387430191,0.23703702725470066],"occluded":false},{"timestamp":3.64,"box":[0.2317708432674408,0.03055555559694767,0.5625000298023224,0.2546296212822199],"occluded":false},{"timestamp":3.9,"box":[0.21927084028720856,0.03055555559694767,0.5500000268220901,0.2703703735023737],"occluded":false},{"timestamp":4.16,"box":[0.1536458432674408,0.03055555559694767,0.5500000417232513,0.2824074123054743],"occluded":false},{"timestamp":4.42,"box":[0.09531250596046448,0.029629629105329514,0.5838541984558105,0.32870371267199516],"occluded":false},{"timestamp":4.68,"box":[0.19583334028720856,0.031481482088565826,0.9911458641290665,0.7481481209397316],"occluded":false},{"timestamp":4.94,"box":[0.30885419249534607,0.3027777671813965,0.7734375596046448,0.7527777552604675],"occluded":false},{"timestamp":5.2,"box":[0.36145836114883423,0.5064814686775208,0.8692708611488342,0.930555522441864],"occluded":false},{"timestamp":5.46,"box":[0.3854166865348816,0.5361111164093018,0.9770833849906921,0.9703703820705414],"occluded":false},{"timestamp":5.72,"box":[0.38697919249534607,0.5740740895271301,0.999479204416275,0.9712963104248047],"occluded":false},{"timestamp":5.98,"box":[0.37447917461395264,0.595370352268219,0.9994792342185974,0.9722222089767456],"occluded":false},{"timestamp":6.24,"box":[0.3760416805744171,0.5972222089767456,0.999479204416275,0.9722222089767456],"occluded":false},{"timestamp":6.5,"box":[0.3697916865348816,0.5907407402992249,0.9994792342185974,0.9712962806224823],"occluded":false},{"timestamp":6.76,"box":[0.36250001192092896,0.5277777910232544,0.9994792342185974,0.9712963104248047],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.3125000298023224,0.5120370388031006,0.999479204416275,0.9712962806224823],"occluded":false},{"timestamp":7.28,"box":[0.20364584028720856,0.4870370328426361,0.8578125387430191,0.9722222089767456],"occluded":false},{"timestamp":7.54,"box":[0.11093750596046448,0.402777761220932,0.7557292282581329,0.9722221791744232],"occluded":false},{"timestamp":7.8,"box":[0.05260416865348816,0.32499998807907104,0.6937500536441803,0.970370352268219],"occluded":false},{"timestamp":8.06,"box":[0.04062500223517418,0.27129629254341125,0.6770833395421505,0.9703703820705414],"occluded":false},{"timestamp":8.32,"box":[0.08906250447034836,0.2768518328666687,0.7046875283122063,0.9694443941116333],"occluded":false},{"timestamp":8.58,"box":[0.1875000149011612,0.38333332538604736,0.8187500387430191,0.970370352268219],"occluded":false},{"timestamp":8.84,"box":[0.31302085518836975,0.41111111640930176,0.9666667282581329,0.970370352268219],"occluded":false},{"timestamp":9.1,"box":[0.4203125238418579,0.29629629850387573,0.9994792342185974,0.8185185194015503],"occluded":false},{"timestamp":9.36,"box":[0.6244791746139526,0.029629629105329514,0.999479204416275,0.46388889476656914],"occluded":false}],"mean_score":0.8947368264198303}],"dropped":{},"fates":[{"noun_phrase":"balcony","object_id":0,"frames":37,"mean_score":0.8947368264198303,"mean_area":0.2351740752521594,"outcome":"kept","presentation_id":3,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"balcony","object_id":1,"frames":7,"mean_score":0.7299578189849854,"mean_area":0.0060365687751184815,"outcome":"kept","presentation_id":1,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"white van","object_id":0,"frames":5,"mean_score":0.6952381134033203,"mean_area":0.012590664238611483,"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/tmp/rmot_smoke_real_86260/renders/--cif7Nv6BI_chunk_011_scene_019_choices.mp4","media_bytes":3819139,"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","temperature":0.0,"as_video":true,"media_bytes":3819139,"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 balcony with the green metal railing on the upper floor\n2. the white van parked on the street\n3. the red car parked on the street","questions":["the balcony with the green metal railing on the upper floor","the white van parked on the street","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\": 1,\n      \"end_frame\": 10,\n      \"confidence\": 1.0\n    },\n    {\n      \"description_index\": 2,\n      \"matching_ids\": [\n        2\n      ],\n      \"start_frame\": 11,\n      \"end_frame\": 12,\n      \"confidence\": 1.0\n    },\n    {\n      \"description_index\": 3,\n      \"matching_ids\": [\n        1\n      ],\n      \"start_frame\": 4,\n      \"end_frame\": 5,\n      \"confidence\": 1.0\n    }\n  ]\n}","raw_answers":[{"description_index":1,"matching_ids":[3],"start_frame":1,"end_frame":10,"confidence":1.0},{"description_index":2,"matching_ids":[2],"start_frame":11,"end_frame":12,"confidence":1.0},{"description_index":3,"matching_ids":[1],"start_frame":4,"end_frame":5,"confidence":1.0}],"answers":[{"expression":"the balcony with the green metal railing on the upper floor","matching_presentation_ids":[3],"start_choice_frame":1,"end_choice_frame":10,"confidence":1.0,"self_reported_confidence":1.0},{"expression":"the white van parked on the street","matching_presentation_ids":[2],"start_choice_frame":11,"end_choice_frame":12,"confidence":1.0,"self_reported_confidence":1.0},{"expression":"the red car parked on the street","matching_presentation_ids":[1],"start_choice_frame":4,"end_choice_frame":5,"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 balcony with the green metal railing on the upper floor","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":[[0.6093750596046448,0.125,0.9994792342185974,0.6083333194255829],[0.596875011920929,0.12777777016162872,0.999479204416275,0.5999999791383743],[0.5765625238418579,0.1509259194135666,0.999479204416275,0.5879629403352737],[0.557812511920929,0.1814814805984497,0.999479204416275,0.5768518447875977],[0.5244792103767395,0.22962962090969086,0.9994792342185974,0.5796296149492264],[0.4828125238418579,0.21111111342906952,0.9994792342185974,0.5898147970438004],[0.4281250238418579,0.14722222089767456,0.9171875417232513,0.567592591047287],[0.37187501788139343,0.07129629701375961,0.8317708671092987,0.5259259268641472],[0.31041666865348816,0.03055555559694767,0.7604166865348816,0.4648148212581873],[0.2510416805744171,0.029629629105329514,0.7067708671092987,0.4018518440425396],[0.19427084922790527,0.029629629105329514,0.6671875417232513,0.3481481485068798],[0.15677084028720856,0.029629629105329514,0.6390625387430191,0.2888888828456402],[0.18281251192092896,0.029629629105329514,0.6151041984558105,0.24629629775881767],[0.1953125149011612,0.03055555559694767,0.5843750387430191,0.23703702725470066],[0.2317708432674408,0.03055555559694767,0.5625000298023224,0.2546296212822199],[0.21927084028720856,0.03055555559694767,0.5500000268220901,0.2703703735023737],[0.1536458432674408,0.03055555559694767,0.5500000417232513,0.2824074123054743],[0.09531250596046448,0.029629629105329514,0.5838541984558105,0.32870371267199516],[0.19583334028720856,0.031481482088565826,0.9911458641290665,0.7481481209397316],[0.30885419249534607,0.3027777671813965,0.7734375596046448,0.7527777552604675],[0.36145836114883423,0.5064814686775208,0.8692708611488342,0.930555522441864],[0.3854166865348816,0.5361111164093018,0.9770833849906921,0.9703703820705414],[0.38697919249534607,0.5740740895271301,0.999479204416275,0.9712963104248047],[0.37447917461395264,0.595370352268219,0.9994792342185974,0.9722222089767456],[0.3760416805744171,0.5972222089767456,0.999479204416275,0.9722222089767456],[0.3697916865348816,0.5907407402992249,0.9994792342185974,0.9712962806224823],[0.36250001192092896,0.5277777910232544,0.9994792342185974,0.9712963104248047],[0.3125000298023224,0.5120370388031006,0.999479204416275,0.9712962806224823],[0.20364584028720856,0.4870370328426361,0.8578125387430191,0.9722222089767456],[0.11093750596046448,0.402777761220932,0.7557292282581329,0.9722221791744232],[0.05260416865348816,0.32499998807907104,0.6937500536441803,0.970370352268219],[0.04062500223517418,0.27129629254341125,0.6770833395421505,0.9703703820705414],[0.08906250447034836,0.2768518328666687,0.7046875283122063,0.9694443941116333],[0.1875000149011612,0.38333332538604736,0.8187500387430191,0.970370352268219],[0.31302085518836975,0.41111111640930176,0.9666667282581329,0.970370352268219],[0.4203125238418579,0.29629629850387573,0.9994792342185974,0.8185185194015503],[0.6244791746139526,0.029629629105329514,0.999479204416275,0.46388889476656914]]}]},{"expression":"the white van parked on the street","kind":"spatial","is_negative":false,"tracks":[{"timestamps":[10.4,10.66,10.92,11.18,11.44],"boxes":[[0.2395833432674408,0.5564814805984497,0.3333333507180214,0.706481471657753],[0.28072917461395264,0.539814829826355,0.369791679084301,0.6777777969837189],[0.30156251788139343,0.5027777552604675,0.3906250223517418,0.6379629373550415],[0.31822916865348816,0.5166666507720947,0.4078125059604645,0.6537036895751953],[0.33020836114883423,0.5259259343147278,0.4192708656191826,0.6638889014720917]]}]}],"rejected":{"the red car parked on the 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_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 balcony with the green metal railing on the upper floor","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":[[0.6093750596046448,0.125,0.9994792342185974,0.6083333194255829],[0.596875011920929,0.12777777016162872,0.999479204416275,0.5999999791383743],[0.5765625238418579,0.1509259194135666,0.999479204416275,0.5879629403352737],[0.557812511920929,0.1814814805984497,0.999479204416275,0.5768518447875977],[0.5244792103767395,0.22962962090969086,0.9994792342185974,0.5796296149492264],[0.4828125238418579,0.21111111342906952,0.9994792342185974,0.5898147970438004],[0.4281250238418579,0.14722222089767456,0.9171875417232513,0.567592591047287],[0.37187501788139343,0.07129629701375961,0.8317708671092987,0.5259259268641472],[0.31041666865348816,0.03055555559694767,0.7604166865348816,0.4648148212581873],[0.2510416805744171,0.029629629105329514,0.7067708671092987,0.4018518440425396],[0.19427084922790527,0.029629629105329514,0.6671875417232513,0.3481481485068798],[0.15677084028720856,0.029629629105329514,0.6390625387430191,0.2888888828456402],[0.18281251192092896,0.029629629105329514,0.6151041984558105,0.24629629775881767],[0.1953125149011612,0.03055555559694767,0.5843750387430191,0.23703702725470066],[0.2317708432674408,0.03055555559694767,0.5625000298023224,0.2546296212822199],[0.21927084028720856,0.03055555559694767,0.5500000268220901,0.2703703735023737],[0.1536458432674408,0.03055555559694767,0.5500000417232513,0.2824074123054743],[0.09531250596046448,0.029629629105329514,0.5838541984558105,0.32870371267199516],[0.19583334028720856,0.031481482088565826,0.9911458641290665,0.7481481209397316],[0.30885419249534607,0.3027777671813965,0.7734375596046448,0.7527777552604675],[0.36145836114883423,0.5064814686775208,0.8692708611488342,0.930555522441864],[0.3854166865348816,0.5361111164093018,0.9770833849906921,0.9703703820705414],[0.38697919249534607,0.5740740895271301,0.999479204416275,0.9712963104248047],[0.37447917461395264,0.595370352268219,0.9994792342185974,0.9722222089767456],[0.3760416805744171,0.5972222089767456,0.999479204416275,0.9722222089767456],[0.3697916865348816,0.5907407402992249,0.9994792342185974,0.9712962806224823],[0.36250001192092896,0.5277777910232544,0.9994792342185974,0.9712963104248047],[0.3125000298023224,0.5120370388031006,0.999479204416275,0.9712962806224823],[0.20364584028720856,0.4870370328426361,0.8578125387430191,0.9722222089767456],[0.11093750596046448,0.402777761220932,0.7557292282581329,0.9722221791744232],[0.05260416865348816,0.32499998807907104,0.6937500536441803,0.970370352268219],[0.04062500223517418,0.27129629254341125,0.6770833395421505,0.9703703820705414],[0.08906250447034836,0.2768518328666687,0.7046875283122063,0.9694443941116333],[0.1875000149011612,0.38333332538604736,0.8187500387430191,0.970370352268219],[0.31302085518836975,0.41111111640930176,0.9666667282581329,0.970370352268219],[0.4203125238418579,0.29629629850387573,0.9994792342185974,0.8185185194015503],[0.6244791746139526,0.029629629105329514,0.999479204416275,0.46388889476656914]]}]},{"expression":"the white van parked on the street","kind":"spatial","is_negative":false,"tracks":[{"timestamps":[10.4,10.66,10.92,11.18,11.44],"boxes":[[0.2395833432674408,0.5564814805984497,0.3333333507180214,0.706481471657753],[0.28072917461395264,0.539814829826355,0.369791679084301,0.6777777969837189],[0.30156251788139343,0.5027777552604675,0.3906250223517418,0.6379629373550415],[0.31822916865348816,0.5166666507720947,0.4078125059604645,0.6537036895751953],[0.33020836114883423,0.5259259343147278,0.4192708656191826,0.6638889014720917]]}]}],"rmot_expression_count":2,"rmot_negative_count":0,"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","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 man with dreadlocks who walks towards the stone building\",\n      \"noun_phrase\": \"man with dreadlocks\",\n      \"kind\": \"motion\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the stone trough filled with water\",\n      \"noun_phrase\": \"stone trough\",\n      \"kind\": \"appearance\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the dry stone trough in the foreground\",\n      \"noun_phrase\": \"stone trough\",\n      \"kind\": \"spatial\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    }\n  ]\n}","raw_items":[{"expression":"the man with dreadlocks who walks towards the stone building","noun_phrase":"man with dreadlocks","kind":"motion","expected_count":1,"is_negative":false},{"expression":"the stone trough filled with water","noun_phrase":"stone trough","kind":"appearance","expected_count":1,"is_negative":false},{"expression":"the dry stone trough in the foreground","noun_phrase":"stone trough","kind":"spatial","expected_count":1,"is_negative":false}],"accepted":[{"expression":"the man with dreadlocks who walks towards the stone building","noun_phrase":"man with dreadlocks","kind":"motion","expected_count":1,"is_negative":false},{"expression":"the stone trough filled with water","noun_phrase":"stone trough","kind":"appearance","expected_count":1,"is_negative":false},{"expression":"the dry stone trough in the foreground","noun_phrase":"stone trough","kind":"spatial","expected_count":1,"is_negative":false}],"gated":[]}},"track":{"inputs":{"tracker":"Sam31MultiplexTracker","model_id":"/mnt/ceph3/ec/jacky/sam3.1","noun_phrases":["man with dreadlocks","stone trough"],"expressions_in":3,"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":"man with dreadlocks","object_id":0,"observations":[{"timestamp":0.52,"box":[0.581250011920929,0.5361111164093018,0.622916679829359,0.8287037014961243],"occluded":false},{"timestamp":0.78,"box":[0.5937500596046448,0.5185185074806213,0.6359375603497028,0.817592591047287],"occluded":false},{"timestamp":1.04,"box":[0.5625,0.5129629373550415,0.6130208373069763,0.8305555284023285],"occluded":false},{"timestamp":1.3,"box":[0.5291666984558105,0.4981481432914734,0.5802083685994148,0.8296296298503876],"occluded":false},{"timestamp":1.56,"box":[0.5453125238418579,0.4851851761341095,0.6005208604037762,0.8388888835906982],"occluded":false},{"timestamp":1.82,"box":[0.5385416746139526,0.49444442987442017,0.5979166775941849,0.8657407164573669],"occluded":false},{"timestamp":2.08,"box":[0.48593753576278687,0.47777777910232544,0.5458333715796471,0.864814817905426],"occluded":false},{"timestamp":2.34,"box":[0.47083336114883423,0.470370352268219,0.5375000312924385,0.8907407224178314],"occluded":false},{"timestamp":2.6,"box":[0.4645833671092987,0.4749999940395355,0.539062537252903,0.9342592358589172],"occluded":false},{"timestamp":2.86,"box":[0.4593750238418579,0.4120370149612427,0.543750025331974,0.8999999761581421],"occluded":false},{"timestamp":3.12,"box":[0.43281251192092896,0.43611109256744385,0.5296875163912773,0.9712962508201599],"occluded":false},{"timestamp":3.38,"box":[0.40156251192092896,0.4120370149612427,0.4984375163912773,0.9694443941116333],"occluded":false},{"timestamp":3.64,"box":[0.40833336114883423,0.42407405376434326,0.49531253427267075,0.9722222089767456],"occluded":false},{"timestamp":3.9,"box":[0.4411458671092987,0.3574073910713196,0.5338542088866234,0.9731481075286865],"occluded":false},{"timestamp":4.16,"box":[0.4177083671092987,0.35555553436279297,0.5286458730697632,0.9712962508201599],"occluded":false},{"timestamp":4.42,"box":[0.3880208432674408,0.29629629850387573,0.526562511920929,0.9722222089767456],"occluded":false},{"timestamp":4.68,"box":[0.3421875238418579,0.22499999403953552,0.5364583730697632,0.9712962806224823],"occluded":false},{"timestamp":4.94,"box":[0.26354166865348816,0.04537037014961243,0.5432291924953461,0.9712962806224823],"occluded":false},{"timestamp":5.2,"box":[0.10104167461395264,0.032407406717538834,0.4921875298023224,0.9722222127020359],"occluded":false},{"timestamp":5.46,"box":[0.0,0.028703702613711357,0.47447919845581055,0.9749999772757292],"occluded":false},{"timestamp":5.72,"box":[0.0,0.028703702613711357,0.4401041865348816,0.9731481205672026],"occluded":false},{"timestamp":5.98,"box":[0.0,0.00277777761220932,0.3333333432674408,0.9722222313284874],"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":[0.0,0.4194444417953491,0.2630208432674408,0.9712963104248047],"occluded":false},{"timestamp":13.26,"box":[0.0,0.11388888955116272,0.2682291865348816,0.9712962806224823],"occluded":false},{"timestamp":13.52,"box":[0.0,0.028703702613711357,0.2640625238418579,0.8990740310400724],"occluded":false},{"timestamp":13.780000000000001,"box":[0.0,0.02777777798473835,0.24687501788139343,0.7861111033707857],"occluded":false},{"timestamp":14.040000000000001,"box":[0.0,0.02777777798473835,0.22812500596046448,0.749074088409543],"occluded":false},{"timestamp":14.3,"box":[0.0,0.02777777798473835,0.2265625149011612,0.7916666734963655],"occluded":false},{"timestamp":14.56,"box":[0.0,0.02777777798473835,0.22552084922790527,0.7861111033707857],"occluded":false},{"timestamp":14.82,"box":[0.0,0.02777777798473835,0.23750001192092896,0.7907407153397799],"occluded":false},{"timestamp":15.08,"box":[0.0,0.026851851493120193,0.24375000596046448,0.7953703589737415],"occluded":false},{"timestamp":15.34,"box":[0.0,0.02777777798473835,0.24687501788139343,0.8259259331971407],"occluded":false},{"timestamp":15.6,"box":[0.0,0.028703702613711357,0.2161458432674408,0.8925925623625517],"occluded":false},{"timestamp":15.860000000000001,"box":[0.0,0.635185182094574,0.06979166716337204,0.9638888835906982],"occluded":false},{"timestamp":17.42,"box":[0.0,0.8685185313224792,0.013541667722165585,0.9722222313284874],"occluded":false},{"timestamp":17.68,"box":[0.0,0.5157407522201538,0.11614584177732468,0.972222238779068],"occluded":false},{"timestamp":17.94,"box":[0.0,0.1675925850868225,0.18541668355464935,0.950925886631012],"occluded":false},{"timestamp":18.2,"box":[0.0,0.046296294778585434,0.20208334922790527,0.9314814768731594],"occluded":false},{"timestamp":18.46,"box":[0.0,0.03055555559694767,0.2109375149011612,0.9231481049209833],"occluded":false},{"timestamp":18.72,"box":[0.0,0.03055555559694767,0.20781251788139343,0.9212963078171015],"occluded":false},{"timestamp":18.98,"box":[0.0,0.032407406717538834,0.20677083730697632,0.941666666418314],"occluded":false},{"timestamp":19.240000000000002,"box":[0.0,0.031481482088565826,0.20781251788139343,0.9370370283722878],"occluded":false},{"timestamp":19.5,"box":[0.0,0.029629629105329514,0.20677083730697632,0.9351851753890514],"occluded":false},{"timestamp":19.76,"box":[0.0,0.029629629105329514,0.20781251788139343,0.9490740709006786],"occluded":false},{"timestamp":20.02,"box":[0.0,0.03055555559694767,0.2083333432674408,0.9601851794868708],"occluded":false},{"timestamp":20.28,"box":[0.0,0.029629629105329514,0.2057291716337204,0.9638888649642467],"occluded":false},{"timestamp":20.54,"box":[0.0,0.028703702613711357,0.26875001192092896,0.9324073921889067],"occluded":false},{"timestamp":20.8,"box":[0.0,0.029629629105329514,0.18958334624767303,0.8888888768851757],"occluded":false},{"timestamp":21.06,"box":[0.0,0.10462962836027145,0.21041667461395264,0.8592592403292656],"occluded":false},{"timestamp":21.32,"box":[0.0,0.520370364189148,0.10000000894069672,0.9314814805984497],"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]},{"noun_phrase":"stone trough","object_id":0,"observations":[{"timestamp":5.46,"box":[0.44322919845581055,0.5462962985038757,0.5369792059063911,0.7175925970077515],"occluded":false},{"timestamp":5.72,"box":[0.44322919845581055,0.5731481313705444,0.5692708790302277,0.7490740567445755],"occluded":false},{"timestamp":5.98,"box":[0.4270833432674408,0.6064814925193787,0.5593750178813934,0.7962962985038757],"occluded":false},{"timestamp":6.24,"box":[0.3812500238418579,0.5944444537162781,0.520833358168602,0.7962962985038757],"occluded":false},{"timestamp":6.5,"box":[0.3671875298023224,0.5999999642372131,0.5109375417232513,0.8175925463438034],"occluded":false},{"timestamp":6.76,"box":[0.39479169249534607,0.5546296238899231,0.546354204416275,0.7824074029922485],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.4515625238418579,0.539814829826355,0.6057291924953461,0.7777777910232544],"occluded":false},{"timestamp":7.28,"box":[0.4776041805744171,0.5148147940635681,0.6432291865348816,0.7648147940635681],"occluded":false},{"timestamp":7.54,"box":[0.4713541865348816,0.5046296119689941,0.6447916924953461,0.7842592298984528],"occluded":false},{"timestamp":7.8,"box":[0.49531251192092896,0.5231481194496155,0.6942708492279053,0.8287036716938019],"occluded":false},{"timestamp":8.06,"box":[0.5479167103767395,0.4851851761341095,0.778125062584877,0.8296296000480652],"occluded":false},{"timestamp":8.32,"box":[0.5442708730697632,0.5777777433395386,0.8135417103767395,0.9666666388511658],"occluded":false},{"timestamp":8.58,"box":[0.5401042103767395,0.5407407283782959,0.8364583849906921,0.9685184955596924],"occluded":false},{"timestamp":8.84,"box":[0.4958333671092987,0.4749999940395355,0.8083333969116211,0.9361110925674438],"occluded":false},{"timestamp":9.1,"box":[0.4932291805744171,0.46759259700775146,0.8479166924953461,0.9722222089767456],"occluded":false},{"timestamp":9.36,"box":[0.5213541984558105,0.39629629254341125,0.9447917342185974,0.9703703820705414],"occluded":false},{"timestamp":9.620000000000001,"box":[0.5458333492279053,0.3870370388031006,0.999479204416275,0.9712963104248047],"occluded":false},{"timestamp":9.88,"box":[0.5390625,0.20462962985038757,0.999479204416275,0.9675925672054291],"occluded":false},{"timestamp":10.14,"box":[0.5020833611488342,0.10925925523042679,0.9994792342185974,0.9222222045063972],"occluded":false},{"timestamp":10.4,"box":[0.49531251192092896,0.07407407462596893,0.9994792342185974,0.963888868689537],"occluded":false},{"timestamp":10.66,"box":[0.5161458849906921,0.031481482088565826,0.9994792342185974,0.8601851835846901],"occluded":false},{"timestamp":10.92,"box":[0.5411458611488342,0.029629629105329514,0.999479204416275,0.7731481604278088],"occluded":false},{"timestamp":11.18,"box":[0.5494791865348816,0.03055555559694767,0.999479204416275,0.7509259078651667],"occluded":false},{"timestamp":11.44,"box":[0.5546875,0.03055555559694767,0.999479204416275,0.7999999616295099],"occluded":false},{"timestamp":11.700000000000001,"box":[0.5614583492279053,0.031481482088565826,0.999479204416275,0.8518518581986427],"occluded":false},{"timestamp":11.96,"box":[0.5739583373069763,0.031481482088565826,0.999479204416275,0.871296264231205],"occluded":false},{"timestamp":12.22,"box":[0.5802083611488342,0.060185182839632034,0.999479204416275,0.9231481440365314],"occluded":false},{"timestamp":12.48,"box":[0.5817708373069763,0.1527777761220932,0.999479204416275,0.9685184806585312],"occluded":false},{"timestamp":12.74,"box":[0.581250011920929,0.13055555522441864,0.999479204416275,0.963888868689537],"occluded":false},{"timestamp":13.0,"box":[0.5770833492279053,0.0962962955236435,0.999479204416275,0.9370370358228683],"occluded":false},{"timestamp":13.26,"box":[0.5776041746139526,0.054629627615213394,0.999479204416275,0.9083333052694798],"occluded":false},{"timestamp":13.52,"box":[0.5755208730697632,0.031481482088565826,0.9994792342185974,0.8749999776482582],"occluded":false},{"timestamp":13.780000000000001,"box":[0.5703125,0.03055555559694767,0.9994791746139526,0.7555555198341608],"occluded":false},{"timestamp":14.040000000000001,"box":[0.550000011920929,0.03055555559694767,0.999479204416275,0.7768518421798944],"occluded":false},{"timestamp":14.3,"box":[0.5520833730697632,0.03055555559694767,0.9994792342185974,0.7879629228264093],"occluded":false},{"timestamp":14.56,"box":[0.5651041865348816,0.03055555559694767,0.999479204416275,0.8249999973922968],"occluded":false},{"timestamp":14.82,"box":[0.5708333849906921,0.03055555559694767,0.9994792342185974,0.8296296093612909],"occluded":false},{"timestamp":15.08,"box":[0.5713542103767395,0.03055555559694767,0.9994792342185974,0.8379629347473383],"occluded":false},{"timestamp":15.34,"box":[0.565625011920929,0.03055555559694767,0.999479204416275,0.8407407496124506],"occluded":false},{"timestamp":15.6,"box":[0.5281250476837158,0.031481482088565826,0.9994792342185974,0.8148147836327553],"occluded":false},{"timestamp":15.860000000000001,"box":[0.12968750298023224,0.032407406717538834,0.9994791895151138,0.9527777470648289],"occluded":false},{"timestamp":16.12,"box":[0.30677086114883423,0.11296296119689941,0.9041666984558105,0.8879629373550415],"occluded":false},{"timestamp":16.38,"box":[0.28125,0.1527777761220932,0.8640625476837158,0.963888868689537],"occluded":false},{"timestamp":16.64,"box":[0.26770836114883423,0.18611110746860504,0.8489583730697632,0.9685185104608536],"occluded":false},{"timestamp":16.9,"box":[0.2734375,0.182407408952713,0.8562500476837158,0.9703703820705414],"occluded":false},{"timestamp":17.16,"box":[0.2968750298023224,0.15555554628372192,0.8911459147930145,0.9657407402992249],"occluded":false},{"timestamp":17.42,"box":[0.3552083373069763,0.0824074074625969,0.9942708611488342,0.9148148223757744],"occluded":false},{"timestamp":17.68,"box":[0.4307291805744171,0.03055555559694767,0.999479204416275,0.914814779534936],"occluded":false},{"timestamp":17.94,"box":[0.48750001192092896,0.031481482088565826,0.9994792342185974,0.8870370164513588],"occluded":false},{"timestamp":18.2,"box":[0.5088542103767395,0.031481482088565826,0.9994792342185974,0.8518518581986427],"occluded":false},{"timestamp":18.46,"box":[0.5145833492279053,0.031481482088565826,0.999479204416275,0.8870370164513588],"occluded":false},{"timestamp":18.72,"box":[0.5104166865348816,0.03055555559694767,0.999479204416275,0.8592592570930719],"occluded":false},{"timestamp":18.98,"box":[0.5437500476837158,0.03055555559694767,0.9994792342185974,0.40462962724268436],"occluded":false},{"timestamp":19.240000000000002,"box":[0.6296875476837158,0.031481482088565826,0.9994792342185974,0.40833333879709244],"occluded":false},{"timestamp":19.5,"box":[0.6338542103767395,0.03055555559694767,0.9994792342185974,0.4092592392116785],"occluded":false},{"timestamp":19.76,"box":[0.6234375238418579,0.03055555559694767,0.999479204416275,0.4222222063690424],"occluded":false},{"timestamp":20.02,"box":[0.6182292103767395,0.031481482088565826,0.9994792342185974,0.43611110001802444],"occluded":false},{"timestamp":20.28,"box":[0.6083333492279053,0.031481482088565826,0.999479204416275,0.4324073866009712],"occluded":false},{"timestamp":20.54,"box":[0.6270833611488342,0.03055555559694767,0.999479204416275,0.4074074123054743],"occluded":false},{"timestamp":20.8,"box":[0.6640625596046448,0.03055555559694767,0.9994792342185974,0.3379629645496607],"occluded":false},{"timestamp":21.06,"box":[0.6364583373069763,0.03055555559694767,0.999479204416275,0.33333332277834415],"occluded":false},{"timestamp":21.32,"box":[0.518750011920929,0.03055555559694767,0.999479204416275,0.5259259138256311],"occluded":false},{"timestamp":21.580000000000002,"box":[0.48750001192092896,0.03055555559694767,0.8401041924953461,0.22685184516012669],"occluded":false},{"timestamp":21.84,"box":[0.15416666865348816,0.031481482088565826,0.5604166984558105,0.20185185223817825],"occluded":false},{"timestamp":22.1,"box":[0.0,0.12314814329147339,0.17239584028720856,0.3083333224058151],"occluded":false},{"timestamp":23.400000000000002,"box":[0.0,0.17129629850387573,0.2588541805744171,0.364814817905426],"occluded":false},{"timestamp":23.66,"box":[0.0,0.046296294778585434,0.550000011920929,0.28333332762122154],"occluded":false},{"timestamp":23.92,"box":[0.12656250596046448,0.03055555559694767,0.8140625655651093,0.27500000037252903],"occluded":false},{"timestamp":24.18,"box":[0.5234375,0.031481482088565826,0.8843750059604645,0.31111110001802444],"occluded":false},{"timestamp":24.44,"box":[0.581250011920929,0.031481482088565826,0.8411458730697632,0.2749999985098839],"occluded":false},{"timestamp":24.7,"box":[0.6822916865348816,0.031481482088565826,0.8963541984558105,0.23888888210058212],"occluded":false},{"timestamp":24.96,"box":[0.745312511920929,0.03055555559694767,0.9244791865348816,0.15000000037252903],"occluded":false},{"timestamp":25.22,"box":[0.7604166865348816,0.031481482088565826,0.8942708671092987,0.10277777910232544],"occluded":false},{"timestamp":26.52,"box":[0.7541667222976685,0.03055555559694767,0.8734375610947609,0.11296296305954456],"occluded":false},{"timestamp":26.78,"box":[0.7307292222976685,0.03055555559694767,0.9572917371988297,0.24074074067175388],"occluded":false},{"timestamp":27.04,"box":[0.7015625238418579,0.031481482088565826,0.9848958849906921,0.32407406717538834],"occluded":false},{"timestamp":27.3,"box":[0.6098958849906921,0.031481482088565826,0.9458334147930145,0.40740741044282913],"occluded":false},{"timestamp":27.560000000000002,"box":[0.5494791865348816,0.05648148059844971,0.9177083671092987,0.44722220301628113],"occluded":false},{"timestamp":27.82,"box":[0.5098958611488342,0.11388888955116272,0.8604167103767395,0.4333333373069763],"occluded":false},{"timestamp":28.080000000000002,"box":[0.4989583492279053,0.18796296417713165,0.835416704416275,0.45277778804302216],"occluded":false},{"timestamp":28.34,"box":[0.4911458492279053,0.23888888955116272,0.8187500238418579,0.510185182094574],"occluded":false},{"timestamp":28.6,"box":[0.47812503576278687,0.28333333134651184,0.8338542282581329,0.5518518388271332],"occluded":false},{"timestamp":28.86,"box":[0.45520836114883423,0.3046296238899231,0.8395833671092987,0.5833333134651184],"occluded":false},{"timestamp":29.12,"box":[0.4505208432674408,0.3092592656612396,0.8515625298023224,0.5972222089767456],"occluded":false},{"timestamp":29.38,"box":[0.4046875238418579,0.307407408952713,0.8098958730697632,0.5435185134410858],"occluded":false}],"scores":[0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349,0.6521739363670349]},{"noun_phrase":"stone trough","object_id":1,"observations":[{"timestamp":5.72,"box":[0.49375003576278687,0.7490740418434143,0.6072917059063911,0.8416666314005852],"occluded":false},{"timestamp":5.98,"box":[0.4755208492279053,0.8027777671813965,0.598437525331974,0.900925911962986],"occluded":false},{"timestamp":6.24,"box":[0.4333333671092987,0.7916666269302368,0.5640625357627869,0.9009258821606636],"occluded":false},{"timestamp":6.5,"box":[0.4281250238418579,0.8046296238899231,0.5651042014360428,0.9314814805984497],"occluded":false},{"timestamp":6.76,"box":[0.45520836114883423,0.7851851582527161,0.6026041954755783,0.9120370149612427],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.518750011920929,0.789814829826355,0.6765625178813934,0.9333333522081375],"occluded":false},{"timestamp":7.28,"box":[0.5458333492279053,0.7972221970558167,0.7218750268220901,0.9611110836267471],"occluded":false},{"timestamp":7.54,"box":[0.5572916865348816,0.8083333373069763,0.7703125327825546,0.9731481522321701],"occluded":false},{"timestamp":7.8,"box":[0.6104167103767395,0.8583332896232605,0.8260417282581329,0.9731481000781059],"occluded":false},{"timestamp":8.06,"box":[0.6953125596046448,0.8805555105209351,0.947395920753479,0.9749999567866325],"occluded":false}],"scores":[0.7142857313156128,0.7142857313156128,0.7142857313156128,0.7142857313156128,0.7142857313156128,0.7142857313156128,0.7142857313156128,0.7142857313156128,0.7142857313156128,0.7142857313156128]},{"noun_phrase":"stone trough","object_id":2,"observations":[{"timestamp":5.98,"box":[0.33229169249534607,0.6175925731658936,0.46145837008953094,0.8370370119810104],"occluded":false},{"timestamp":6.24,"box":[0.27812501788139343,0.625,0.4177083522081375,0.8490740656852722],"occluded":false},{"timestamp":6.5,"box":[0.2640625238418579,0.6370370388031006,0.4088542014360428,0.8740740716457367],"occluded":false},{"timestamp":6.76,"box":[0.2875000238418579,0.5814814567565918,0.43645836412906647,0.8212962746620178],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.34583336114883423,0.5694444179534912,0.49635420739650726,0.8148147910833359],"occluded":false},{"timestamp":7.28,"box":[0.3661458492279053,0.54537034034729,0.5239583551883698,0.8046295940876007],"occluded":false},{"timestamp":7.54,"box":[0.35572919249534607,0.5333333015441895,0.5244791954755783,0.8240740299224854],"occluded":false},{"timestamp":7.8,"box":[0.38229167461395264,0.5462962985038757,0.5666666775941849,0.8731481432914734],"occluded":false},{"timestamp":8.06,"box":[0.4333333671092987,0.5,0.6348958760499954,0.8601851761341095],"occluded":false},{"timestamp":8.32,"box":[0.4229166805744171,0.5990740656852722,0.6609375327825546,0.9712962806224823],"occluded":false},{"timestamp":8.58,"box":[0.4098958671092987,0.5638889074325562,0.6677083671092987,0.9990740716457367],"occluded":false},{"timestamp":8.84,"box":[0.35989585518836975,0.5074074268341064,0.6458333730697632,0.9712963104248047],"occluded":false},{"timestamp":9.1,"box":[0.34687501192092896,0.49074074625968933,0.6765625178813934,0.9712962806224823],"occluded":false},{"timestamp":9.36,"box":[0.37135419249534607,0.4203703701496124,0.7484375536441803,0.9712962806224823],"occluded":false},{"timestamp":9.620000000000001,"box":[0.38072919845581055,0.42499998211860657,0.9218750596046448,0.9712962806224823],"occluded":false},{"timestamp":9.88,"box":[0.3453125059604645,0.23888888955116272,0.8947916924953461,0.9712962806224823],"occluded":false},{"timestamp":10.14,"box":[0.2770833373069763,0.15555554628372192,0.8296875357627869,0.9712962508201599],"occluded":false},{"timestamp":10.4,"box":[0.2458333522081375,0.12314814329147339,0.9796875864267349,0.9712962508201599],"occluded":false},{"timestamp":10.66,"box":[0.24947917461395264,0.031481482088565826,0.9250000715255737,0.9712962880730629],"occluded":false},{"timestamp":10.92,"box":[0.2421875149011612,0.031481482088565826,0.9489584118127823,0.9703703299164772],"occluded":false},{"timestamp":11.18,"box":[0.24687501788139343,0.031481482088565826,0.9718750417232513,0.9703703299164772],"occluded":false},{"timestamp":11.44,"box":[0.2536458373069763,0.031481482088565826,0.9781250357627869,0.9703703299164772],"occluded":false},{"timestamp":11.700000000000001,"box":[0.26770836114883423,0.031481482088565826,0.9994792342185974,0.9712962880730629],"occluded":false},{"timestamp":11.96,"box":[0.27916666865348816,0.07777777314186096,0.999479204416275,0.9712962806224823],"occluded":false},{"timestamp":12.22,"box":[0.2822916805744171,0.17222221195697784,0.9911458790302277,0.9712962657213211],"occluded":false},{"timestamp":12.48,"box":[0.28177085518836975,0.21574074029922485,0.9828125536441803,0.9712963104248047],"occluded":false},{"timestamp":12.74,"box":[0.28125,0.18981480598449707,0.979687511920929,0.970370352268219],"occluded":false},{"timestamp":13.0,"box":[0.2744791805744171,0.15648147463798523,0.9859375655651093,0.9712962806224823],"occluded":false},{"timestamp":13.26,"box":[0.26875001192092896,0.11203703284263611,0.9875000715255737,0.9712962806224823],"occluded":false},{"timestamp":13.52,"box":[0.25989586114883423,0.032407406717538834,0.9906250834465027,0.9712962545454502],"occluded":false},{"timestamp":13.780000000000001,"box":[0.24062500894069672,0.031481482088565826,0.9718750566244125,0.9703703299164772],"occluded":false},{"timestamp":14.040000000000001,"box":[0.2265625149011612,0.031481482088565826,0.9479167014360428,0.9703703299164772],"occluded":false},{"timestamp":14.3,"box":[0.22343751788139343,0.031481482088565826,0.9468750655651093,0.9712962880730629],"occluded":false},{"timestamp":14.56,"box":[0.22500000894069672,0.031481482088565826,0.9520833939313889,0.9703703299164772],"occluded":false},{"timestamp":14.82,"box":[0.23645834624767303,0.031481482088565826,0.9651042073965073,0.9703703299164772],"occluded":false},{"timestamp":15.08,"box":[0.24270834028720856,0.031481482088565826,0.9713542014360428,0.9703703299164772],"occluded":false},{"timestamp":15.34,"box":[0.24427084624767303,0.031481482088565826,0.9671875685453415,0.9703703299164772],"occluded":false},{"timestamp":15.6,"box":[0.21302084624767303,0.031481482088565826,0.9244792312383652,0.9703703299164772],"occluded":false},{"timestamp":15.860000000000001,"box":[0.04635417088866234,0.06574074178934097,0.7901042066514492,0.9712962880730629],"occluded":false},{"timestamp":16.12,"box":[0.0,0.21296295523643494,0.7031250596046448,0.9712962806224823],"occluded":false},{"timestamp":16.38,"box":[0.0,0.28148147463798523,0.6697916984558105,0.9712962806224823],"occluded":false},{"timestamp":16.64,"box":[0.22708334028720856,0.5935184955596924,0.6656250208616257,0.9712962806224823],"occluded":false},{"timestamp":16.9,"box":[0.3697916865348816,0.5870370268821716,0.6708333790302277,0.9712962806224823],"occluded":false},{"timestamp":17.16,"box":[0.31041666865348816,0.5731481313705444,0.6953125298023224,0.9712962806224823],"occluded":false},{"timestamp":17.42,"box":[0.2682291865348816,0.49259257316589355,0.7588542103767395,0.9712962806224823],"occluded":false},{"timestamp":17.68,"box":[0.11510417610406876,0.08518518507480621,0.835937537252903,0.9712962657213211],"occluded":false},{"timestamp":17.94,"box":[0.17916667461395264,0.032407406717538834,0.9994792342185974,0.9712962545454502],"occluded":false},{"timestamp":18.2,"box":[0.19843751192092896,0.032407406717538834,0.9703125357627869,0.9712962545454502],"occluded":false},{"timestamp":18.46,"box":[0.20260417461395264,0.032407406717538834,0.9994792342185974,0.9712962545454502],"occluded":false},{"timestamp":18.72,"box":[0.2005208432674408,0.032407406717538834,0.999479204416275,0.9712962545454502],"occluded":false}],"scores":[0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458,0.8202247023582458]},{"noun_phrase":"stone trough","object_id":3,"observations":[{"timestamp":19.5,"box":[0.19947917759418488,0.032407406717538834,0.9994792491197586,0.9694443978369236],"occluded":false},{"timestamp":19.76,"box":[0.2005208432674408,0.032407406717538834,0.999479204416275,0.9703703559935093],"occluded":false},{"timestamp":20.02,"box":[0.20208334922790527,0.032407406717538834,0.9890625476837158,0.9703703559935093],"occluded":false},{"timestamp":20.28,"box":[0.2005208432674408,0.032407406717538834,0.9359375536441803,0.9703703559935093],"occluded":false},{"timestamp":20.54,"box":[0.18802084028720856,0.032407406717538834,0.960937574505806,0.9703703559935093],"occluded":false},{"timestamp":20.8,"box":[0.17760416865348816,0.031481482088565826,0.9494791924953461,0.9703703299164772],"occluded":false},{"timestamp":21.06,"box":[0.14739583432674408,0.031481482088565826,0.9994792193174362,0.9703703299164772],"occluded":false},{"timestamp":21.32,"box":[0.0807291716337204,0.032407406717538834,0.9354167431592941,0.9703703559935093],"occluded":false},{"timestamp":21.580000000000002,"box":[0.0,0.031481482088565826,0.9427083730697632,0.9712962880730629],"occluded":false}],"scores":[0.6878048777580261,0.6878048777580261,0.6878048777580261,0.6878048777580261,0.6878048777580261,0.6878048777580261,0.6878048777580261,0.6878048777580261,0.6878048777580261]},{"noun_phrase":"stone trough","object_id":4,"observations":[{"timestamp":22.1,"box":[0.5317708849906921,0.25740739703178406,0.8333334028720856,0.902777761220932],"occluded":false},{"timestamp":22.36,"box":[0.01875000074505806,0.2796296179294586,0.6802083738148212,0.9990740716457367],"occluded":false},{"timestamp":22.62,"box":[0.19947917759418488,0.317592591047287,0.6369792073965073,0.9990740716457367],"occluded":false},{"timestamp":22.88,"box":[0.04635417088866234,0.3305555582046509,0.6421875320374966,0.9990740418434143],"occluded":false},{"timestamp":23.14,"box":[0.2723958492279053,0.287962943315506,0.6661458611488342,0.9564814269542694],"occluded":false},{"timestamp":23.400000000000002,"box":[0.5276042222976685,0.3083333373069763,0.84270840883255,0.9537037014961243],"occluded":false},{"timestamp":23.66,"box":[0.7473958730697632,0.41111111640930176,0.9994792342185974,0.9712963104248047],"occluded":false}],"scores":[0.8827838897705078,0.8827838897705078,0.8827838897705078,0.8827838897705078,0.8827838897705078,0.8827838897705078,0.8827838897705078]},{"noun_phrase":"stone trough","object_id":5,"observations":[{"timestamp":24.18,"box":[0.03958333656191826,0.03333333134651184,0.9994792453944683,0.9694444239139557],"occluded":false},{"timestamp":24.44,"box":[0.10208334028720856,0.032407406717538834,0.9994792491197586,0.9703703559935093],"occluded":false},{"timestamp":24.7,"box":[0.13437500596046448,0.031481482088565826,0.999479204416275,0.9703703299164772],"occluded":false},{"timestamp":24.96,"box":[0.15000000596046448,0.031481482088565826,0.999479204416275,0.9712962880730629],"occluded":false},{"timestamp":25.22,"box":[0.09479167312383652,0.031481482088565826,0.9994791969656944,0.9703703299164772],"occluded":false},{"timestamp":25.48,"box":[0.02239583432674408,0.031481482088565826,0.9994792193174362,0.9712962880730629],"occluded":false},{"timestamp":25.740000000000002,"box":[0.0,0.031481482088565826,0.9994792342185974,0.9703703299164772],"occluded":false},{"timestamp":26.0,"box":[0.0,0.03055555559694767,0.9994792342185974,0.9712962601333857],"occluded":false},{"timestamp":26.26,"box":[0.004687500186264515,0.03055555559694767,0.9994792463257909,0.9703703615814447],"occluded":false},{"timestamp":26.52,"box":[0.05572916939854622,0.031481482088565826,0.9994791932404041,0.9712962880730629],"occluded":false},{"timestamp":26.78,"box":[0.12135417014360428,0.031481482088565826,0.999479241669178,0.9712962880730629],"occluded":false},{"timestamp":27.04,"box":[0.1484375149011612,0.031481482088565826,0.9994792491197586,0.9712962880730629],"occluded":false},{"timestamp":27.3,"box":[0.07395833730697632,0.032407406717538834,0.9994792342185974,0.9712962545454502],"occluded":false},{"timestamp":27.560000000000002,"box":[0.18125000596046448,0.03333333134651184,0.999479204416275,0.9712962806224823],"occluded":false},{"timestamp":27.82,"box":[0.21562501788139343,0.07685185223817825,0.999479204416275,0.9712962582707405],"occluded":false},{"timestamp":28.080000000000002,"box":[0.2447916865348816,0.1509259194135666,0.9994792342185974,0.9722221940755844],"occluded":false},{"timestamp":28.34,"box":[0.24010418355464935,0.20555555820465088,0.9994792193174362,0.9712963104248047],"occluded":false},{"timestamp":28.6,"box":[0.22708334028720856,0.24722221493721008,0.9994791895151138,0.9712962806224823],"occluded":false},{"timestamp":28.86,"box":[0.20208334922790527,0.27129629254341125,0.9994792342185974,0.9712962806224823],"occluded":false},{"timestamp":29.12,"box":[0.18854168057441711,0.277777761220932,0.999479204416275,0.9712962806224823],"occluded":false},{"timestamp":29.38,"box":[0.33802086114883423,0.2787036895751953,0.9994792342185974,0.9694443941116333],"occluded":false}],"scores":[0.7815700173377991,0.7815700173377991,0.7815700173377991,0.7815700173377991,0.7815700173377991,0.7815700173377991,0.7815700173377991,0.7815700173377991,0.7815700173377991,0.7815700173377991,0.7815700173377991,0.7815700173377991,0.7815700173377991,0.7815700173377991,0.7815700173377991,0.7815700173377991,0.7815700173377991,0.7815700173377991,0.7815700173377991,0.7815700173377991,0.7815700173377991]}],"tracklets":[{"noun_phrase":"stone trough","object_id":0,"presentation_id":1,"observations":[{"timestamp":5.46,"box":[0.44322919845581055,0.5462962985038757,0.5369792059063911,0.7175925970077515],"occluded":false},{"timestamp":5.72,"box":[0.44322919845581055,0.5731481313705444,0.5692708790302277,0.7490740567445755],"occluded":false},{"timestamp":5.98,"box":[0.4270833432674408,0.6064814925193787,0.5593750178813934,0.7962962985038757],"occluded":false},{"timestamp":6.24,"box":[0.3812500238418579,0.5944444537162781,0.520833358168602,0.7962962985038757],"occluded":false},{"timestamp":6.5,"box":[0.3671875298023224,0.5999999642372131,0.5109375417232513,0.8175925463438034],"occluded":false},{"timestamp":6.76,"box":[0.39479169249534607,0.5546296238899231,0.546354204416275,0.7824074029922485],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.4515625238418579,0.539814829826355,0.6057291924953461,0.7777777910232544],"occluded":false},{"timestamp":7.28,"box":[0.4776041805744171,0.5148147940635681,0.6432291865348816,0.7648147940635681],"occluded":false},{"timestamp":7.54,"box":[0.4713541865348816,0.5046296119689941,0.6447916924953461,0.7842592298984528],"occluded":false},{"timestamp":7.8,"box":[0.49531251192092896,0.5231481194496155,0.6942708492279053,0.8287036716938019],"occluded":false},{"timestamp":8.06,"box":[0.5479167103767395,0.4851851761341095,0.778125062584877,0.8296296000480652],"occluded":false},{"timestamp":8.32,"box":[0.5442708730697632,0.5777777433395386,0.8135417103767395,0.9666666388511658],"occluded":false},{"timestamp":8.58,"box":[0.5401042103767395,0.5407407283782959,0.8364583849906921,0.9685184955596924],"occluded":false},{"timestamp":8.84,"box":[0.4958333671092987,0.4749999940395355,0.8083333969116211,0.9361110925674438],"occluded":false},{"timestamp":9.1,"box":[0.4932291805744171,0.46759259700775146,0.8479166924953461,0.9722222089767456],"occluded":false},{"timestamp":9.36,"box":[0.5213541984558105,0.39629629254341125,0.9447917342185974,0.9703703820705414],"occluded":false},{"timestamp":9.620000000000001,"box":[0.5458333492279053,0.3870370388031006,0.999479204416275,0.9712963104248047],"occluded":false},{"timestamp":9.88,"box":[0.5390625,0.20462962985038757,0.999479204416275,0.9675925672054291],"occluded":false},{"timestamp":10.14,"box":[0.5020833611488342,0.10925925523042679,0.9994792342185974,0.9222222045063972],"occluded":false},{"timestamp":10.4,"box":[0.49531251192092896,0.07407407462596893,0.9994792342185974,0.963888868689537],"occluded":false},{"timestamp":10.66,"box":[0.5161458849906921,0.031481482088565826,0.9994792342185974,0.8601851835846901],"occluded":false},{"timestamp":10.92,"box":[0.5411458611488342,0.029629629105329514,0.999479204416275,0.7731481604278088],"occluded":false},{"timestamp":11.18,"box":[0.5494791865348816,0.03055555559694767,0.999479204416275,0.7509259078651667],"occluded":false},{"timestamp":11.44,"box":[0.5546875,0.03055555559694767,0.999479204416275,0.7999999616295099],"occluded":false},{"timestamp":11.700000000000001,"box":[0.5614583492279053,0.031481482088565826,0.999479204416275,0.8518518581986427],"occluded":false},{"timestamp":11.96,"box":[0.5739583373069763,0.031481482088565826,0.999479204416275,0.871296264231205],"occluded":false},{"timestamp":12.22,"box":[0.5802083611488342,0.060185182839632034,0.999479204416275,0.9231481440365314],"occluded":false},{"timestamp":12.48,"box":[0.5817708373069763,0.1527777761220932,0.999479204416275,0.9685184806585312],"occluded":false},{"timestamp":12.74,"box":[0.581250011920929,0.13055555522441864,0.999479204416275,0.963888868689537],"occluded":false},{"timestamp":13.0,"box":[0.5770833492279053,0.0962962955236435,0.999479204416275,0.9370370358228683],"occluded":false},{"timestamp":13.26,"box":[0.5776041746139526,0.054629627615213394,0.999479204416275,0.9083333052694798],"occluded":false},{"timestamp":13.52,"box":[0.5755208730697632,0.031481482088565826,0.9994792342185974,0.8749999776482582],"occluded":false},{"timestamp":13.780000000000001,"box":[0.5703125,0.03055555559694767,0.9994791746139526,0.7555555198341608],"occluded":false},{"timestamp":14.040000000000001,"box":[0.550000011920929,0.03055555559694767,0.999479204416275,0.7768518421798944],"occluded":false},{"timestamp":14.3,"box":[0.5520833730697632,0.03055555559694767,0.9994792342185974,0.7879629228264093],"occluded":false},{"timestamp":14.56,"box":[0.5651041865348816,0.03055555559694767,0.999479204416275,0.8249999973922968],"occluded":false},{"timestamp":14.82,"box":[0.5708333849906921,0.03055555559694767,0.9994792342185974,0.8296296093612909],"occluded":false},{"timestamp":15.08,"box":[0.5713542103767395,0.03055555559694767,0.9994792342185974,0.8379629347473383],"occluded":false},{"timestamp":15.34,"box":[0.565625011920929,0.03055555559694767,0.999479204416275,0.8407407496124506],"occluded":false},{"timestamp":15.6,"box":[0.5281250476837158,0.031481482088565826,0.9994792342185974,0.8148147836327553],"occluded":false},{"timestamp":15.860000000000001,"box":[0.12968750298023224,0.032407406717538834,0.9994791895151138,0.9527777470648289],"occluded":false},{"timestamp":16.12,"box":[0.30677086114883423,0.11296296119689941,0.9041666984558105,0.8879629373550415],"occluded":false},{"timestamp":16.38,"box":[0.28125,0.1527777761220932,0.8640625476837158,0.963888868689537],"occluded":false},{"timestamp":16.64,"box":[0.26770836114883423,0.18611110746860504,0.8489583730697632,0.9685185104608536],"occluded":false},{"timestamp":16.9,"box":[0.2734375,0.182407408952713,0.8562500476837158,0.9703703820705414],"occluded":false},{"timestamp":17.16,"box":[0.2968750298023224,0.15555554628372192,0.8911459147930145,0.9657407402992249],"occluded":false},{"timestamp":17.42,"box":[0.3552083373069763,0.0824074074625969,0.9942708611488342,0.9148148223757744],"occluded":false},{"timestamp":17.68,"box":[0.4307291805744171,0.03055555559694767,0.999479204416275,0.914814779534936],"occluded":false},{"timestamp":17.94,"box":[0.48750001192092896,0.031481482088565826,0.9994792342185974,0.8870370164513588],"occluded":false},{"timestamp":18.2,"box":[0.5088542103767395,0.031481482088565826,0.9994792342185974,0.8518518581986427],"occluded":false},{"timestamp":18.46,"box":[0.5145833492279053,0.031481482088565826,0.999479204416275,0.8870370164513588],"occluded":false},{"timestamp":18.72,"box":[0.5104166865348816,0.03055555559694767,0.999479204416275,0.8592592570930719],"occluded":false},{"timestamp":18.98,"box":[0.5437500476837158,0.03055555559694767,0.9994792342185974,0.40462962724268436],"occluded":false},{"timestamp":19.240000000000002,"box":[0.6296875476837158,0.031481482088565826,0.9994792342185974,0.40833333879709244],"occluded":false},{"timestamp":19.5,"box":[0.6338542103767395,0.03055555559694767,0.9994792342185974,0.4092592392116785],"occluded":false},{"timestamp":19.76,"box":[0.6234375238418579,0.03055555559694767,0.999479204416275,0.4222222063690424],"occluded":false},{"timestamp":20.02,"box":[0.6182292103767395,0.031481482088565826,0.9994792342185974,0.43611110001802444],"occluded":false},{"timestamp":20.28,"box":[0.6083333492279053,0.031481482088565826,0.999479204416275,0.4324073866009712],"occluded":false},{"timestamp":20.54,"box":[0.6270833611488342,0.03055555559694767,0.999479204416275,0.4074074123054743],"occluded":false},{"timestamp":20.8,"box":[0.6640625596046448,0.03055555559694767,0.9994792342185974,0.3379629645496607],"occluded":false},{"timestamp":21.06,"box":[0.6364583373069763,0.03055555559694767,0.999479204416275,0.33333332277834415],"occluded":false},{"timestamp":21.32,"box":[0.518750011920929,0.03055555559694767,0.999479204416275,0.5259259138256311],"occluded":false},{"timestamp":21.580000000000002,"box":[0.48750001192092896,0.03055555559694767,0.8401041924953461,0.22685184516012669],"occluded":false},{"timestamp":21.84,"box":[0.15416666865348816,0.031481482088565826,0.5604166984558105,0.20185185223817825],"occluded":false},{"timestamp":22.1,"box":[0.0,0.12314814329147339,0.17239584028720856,0.3083333224058151],"occluded":false},{"timestamp":23.400000000000002,"box":[0.0,0.17129629850387573,0.2588541805744171,0.364814817905426],"occluded":false},{"timestamp":23.66,"box":[0.0,0.046296294778585434,0.550000011920929,0.28333332762122154],"occluded":false},{"timestamp":23.92,"box":[0.12656250596046448,0.03055555559694767,0.8140625655651093,0.27500000037252903],"occluded":false},{"timestamp":24.18,"box":[0.5234375,0.031481482088565826,0.8843750059604645,0.31111110001802444],"occluded":false},{"timestamp":24.44,"box":[0.581250011920929,0.031481482088565826,0.8411458730697632,0.2749999985098839],"occluded":false},{"timestamp":24.7,"box":[0.6822916865348816,0.031481482088565826,0.8963541984558105,0.23888888210058212],"occluded":false},{"timestamp":24.96,"box":[0.745312511920929,0.03055555559694767,0.9244791865348816,0.15000000037252903],"occluded":false},{"timestamp":25.22,"box":[0.7604166865348816,0.031481482088565826,0.8942708671092987,0.10277777910232544],"occluded":false},{"timestamp":26.52,"box":[0.7541667222976685,0.03055555559694767,0.8734375610947609,0.11296296305954456],"occluded":false},{"timestamp":26.78,"box":[0.7307292222976685,0.03055555559694767,0.9572917371988297,0.24074074067175388],"occluded":false},{"timestamp":27.04,"box":[0.7015625238418579,0.031481482088565826,0.9848958849906921,0.32407406717538834],"occluded":false},{"timestamp":27.3,"box":[0.6098958849906921,0.031481482088565826,0.9458334147930145,0.40740741044282913],"occluded":false},{"timestamp":27.560000000000002,"box":[0.5494791865348816,0.05648148059844971,0.9177083671092987,0.44722220301628113],"occluded":false},{"timestamp":27.82,"box":[0.5098958611488342,0.11388888955116272,0.8604167103767395,0.4333333373069763],"occluded":false},{"timestamp":28.080000000000002,"box":[0.4989583492279053,0.18796296417713165,0.835416704416275,0.45277778804302216],"occluded":false},{"timestamp":28.34,"box":[0.4911458492279053,0.23888888955116272,0.8187500238418579,0.510185182094574],"occluded":false},{"timestamp":28.6,"box":[0.47812503576278687,0.28333333134651184,0.8338542282581329,0.5518518388271332],"occluded":false},{"timestamp":28.86,"box":[0.45520836114883423,0.3046296238899231,0.8395833671092987,0.5833333134651184],"occluded":false},{"timestamp":29.12,"box":[0.4505208432674408,0.3092592656612396,0.8515625298023224,0.5972222089767456],"occluded":false},{"timestamp":29.38,"box":[0.4046875238418579,0.307407408952713,0.8098958730697632,0.5435185134410858],"occluded":false}],"mean_score":0.6521739363670349},{"noun_phrase":"man with dreadlocks","object_id":0,"presentation_id":2,"observations":[{"timestamp":0.52,"box":[0.581250011920929,0.5361111164093018,0.622916679829359,0.8287037014961243],"occluded":false},{"timestamp":0.78,"box":[0.5937500596046448,0.5185185074806213,0.6359375603497028,0.817592591047287],"occluded":false},{"timestamp":1.04,"box":[0.5625,0.5129629373550415,0.6130208373069763,0.8305555284023285],"occluded":false},{"timestamp":1.3,"box":[0.5291666984558105,0.4981481432914734,0.5802083685994148,0.8296296298503876],"occluded":false},{"timestamp":1.56,"box":[0.5453125238418579,0.4851851761341095,0.6005208604037762,0.8388888835906982],"occluded":false},{"timestamp":1.82,"box":[0.5385416746139526,0.49444442987442017,0.5979166775941849,0.8657407164573669],"occluded":false},{"timestamp":2.08,"box":[0.48593753576278687,0.47777777910232544,0.5458333715796471,0.864814817905426],"occluded":false},{"timestamp":2.34,"box":[0.47083336114883423,0.470370352268219,0.5375000312924385,0.8907407224178314],"occluded":false},{"timestamp":2.6,"box":[0.4645833671092987,0.4749999940395355,0.539062537252903,0.9342592358589172],"occluded":false},{"timestamp":2.86,"box":[0.4593750238418579,0.4120370149612427,0.543750025331974,0.8999999761581421],"occluded":false},{"timestamp":3.12,"box":[0.43281251192092896,0.43611109256744385,0.5296875163912773,0.9712962508201599],"occluded":false},{"timestamp":3.38,"box":[0.40156251192092896,0.4120370149612427,0.4984375163912773,0.9694443941116333],"occluded":false},{"timestamp":3.64,"box":[0.40833336114883423,0.42407405376434326,0.49531253427267075,0.9722222089767456],"occluded":false},{"timestamp":3.9,"box":[0.4411458671092987,0.3574073910713196,0.5338542088866234,0.9731481075286865],"occluded":false},{"timestamp":4.16,"box":[0.4177083671092987,0.35555553436279297,0.5286458730697632,0.9712962508201599],"occluded":false},{"timestamp":4.42,"box":[0.3880208432674408,0.29629629850387573,0.526562511920929,0.9722222089767456],"occluded":false},{"timestamp":4.68,"box":[0.3421875238418579,0.22499999403953552,0.5364583730697632,0.9712962806224823],"occluded":false},{"timestamp":4.94,"box":[0.26354166865348816,0.04537037014961243,0.5432291924953461,0.9712962806224823],"occluded":false},{"timestamp":5.2,"box":[0.10104167461395264,0.032407406717538834,0.4921875298023224,0.9722222127020359],"occluded":false},{"timestamp":5.46,"box":[0.0,0.028703702613711357,0.47447919845581055,0.9749999772757292],"occluded":false},{"timestamp":5.72,"box":[0.0,0.028703702613711357,0.4401041865348816,0.9731481205672026],"occluded":false},{"timestamp":5.98,"box":[0.0,0.00277777761220932,0.3333333432674408,0.9722222313284874],"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":[0.0,0.4194444417953491,0.2630208432674408,0.9712963104248047],"occluded":false},{"timestamp":13.26,"box":[0.0,0.11388888955116272,0.2682291865348816,0.9712962806224823],"occluded":false},{"timestamp":13.52,"box":[0.0,0.028703702613711357,0.2640625238418579,0.8990740310400724],"occluded":false},{"timestamp":13.780000000000001,"box":[0.0,0.02777777798473835,0.24687501788139343,0.7861111033707857],"occluded":false},{"timestamp":14.040000000000001,"box":[0.0,0.02777777798473835,0.22812500596046448,0.749074088409543],"occluded":false},{"timestamp":14.3,"box":[0.0,0.02777777798473835,0.2265625149011612,0.7916666734963655],"occluded":false},{"timestamp":14.56,"box":[0.0,0.02777777798473835,0.22552084922790527,0.7861111033707857],"occluded":false},{"timestamp":14.82,"box":[0.0,0.02777777798473835,0.23750001192092896,0.7907407153397799],"occluded":false},{"timestamp":15.08,"box":[0.0,0.026851851493120193,0.24375000596046448,0.7953703589737415],"occluded":false},{"timestamp":15.34,"box":[0.0,0.02777777798473835,0.24687501788139343,0.8259259331971407],"occluded":false},{"timestamp":15.6,"box":[0.0,0.028703702613711357,0.2161458432674408,0.8925925623625517],"occluded":false},{"timestamp":15.860000000000001,"box":[0.0,0.635185182094574,0.06979166716337204,0.9638888835906982],"occluded":false},{"timestamp":17.42,"box":[0.0,0.8685185313224792,0.013541667722165585,0.9722222313284874],"occluded":false},{"timestamp":17.68,"box":[0.0,0.5157407522201538,0.11614584177732468,0.972222238779068],"occluded":false},{"timestamp":17.94,"box":[0.0,0.1675925850868225,0.18541668355464935,0.950925886631012],"occluded":false},{"timestamp":18.2,"box":[0.0,0.046296294778585434,0.20208334922790527,0.9314814768731594],"occluded":false},{"timestamp":18.46,"box":[0.0,0.03055555559694767,0.2109375149011612,0.9231481049209833],"occluded":false},{"timestamp":18.72,"box":[0.0,0.03055555559694767,0.20781251788139343,0.9212963078171015],"occluded":false},{"timestamp":18.98,"box":[0.0,0.032407406717538834,0.20677083730697632,0.941666666418314],"occluded":false},{"timestamp":19.240000000000002,"box":[0.0,0.031481482088565826,0.20781251788139343,0.9370370283722878],"occluded":false},{"timestamp":19.5,"box":[0.0,0.029629629105329514,0.20677083730697632,0.9351851753890514],"occluded":false},{"timestamp":19.76,"box":[0.0,0.029629629105329514,0.20781251788139343,0.9490740709006786],"occluded":false},{"timestamp":20.02,"box":[0.0,0.03055555559694767,0.2083333432674408,0.9601851794868708],"occluded":false},{"timestamp":20.28,"box":[0.0,0.029629629105329514,0.2057291716337204,0.9638888649642467],"occluded":false},{"timestamp":20.54,"box":[0.0,0.028703702613711357,0.26875001192092896,0.9324073921889067],"occluded":false},{"timestamp":20.8,"box":[0.0,0.029629629105329514,0.18958334624767303,0.8888888768851757],"occluded":false},{"timestamp":21.06,"box":[0.0,0.10462962836027145,0.21041667461395264,0.8592592403292656],"occluded":false},{"timestamp":21.32,"box":[0.0,0.520370364189148,0.10000000894069672,0.9314814805984497],"occluded":false}],"mean_score":0.6831682920455933},{"noun_phrase":"stone trough","object_id":1,"presentation_id":3,"observations":[{"timestamp":5.72,"box":[0.49375003576278687,0.7490740418434143,0.6072917059063911,0.8416666314005852],"occluded":false},{"timestamp":5.98,"box":[0.4755208492279053,0.8027777671813965,0.598437525331974,0.900925911962986],"occluded":false},{"timestamp":6.24,"box":[0.4333333671092987,0.7916666269302368,0.5640625357627869,0.9009258821606636],"occluded":false},{"timestamp":6.5,"box":[0.4281250238418579,0.8046296238899231,0.5651042014360428,0.9314814805984497],"occluded":false},{"timestamp":6.76,"box":[0.45520836114883423,0.7851851582527161,0.6026041954755783,0.9120370149612427],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.518750011920929,0.789814829826355,0.6765625178813934,0.9333333522081375],"occluded":false},{"timestamp":7.28,"box":[0.5458333492279053,0.7972221970558167,0.7218750268220901,0.9611110836267471],"occluded":false},{"timestamp":7.54,"box":[0.5572916865348816,0.8083333373069763,0.7703125327825546,0.9731481522321701],"occluded":false},{"timestamp":7.8,"box":[0.6104167103767395,0.8583332896232605,0.8260417282581329,0.9731481000781059],"occluded":false},{"timestamp":8.06,"box":[0.6953125596046448,0.8805555105209351,0.947395920753479,0.9749999567866325],"occluded":false}],"mean_score":0.7142857313156128},{"noun_phrase":"stone trough","object_id":4,"presentation_id":4,"observations":[{"timestamp":22.1,"box":[0.5317708849906921,0.25740739703178406,0.8333334028720856,0.902777761220932],"occluded":false},{"timestamp":22.36,"box":[0.01875000074505806,0.2796296179294586,0.6802083738148212,0.9990740716457367],"occluded":false},{"timestamp":22.62,"box":[0.19947917759418488,0.317592591047287,0.6369792073965073,0.9990740716457367],"occluded":false},{"timestamp":22.88,"box":[0.04635417088866234,0.3305555582046509,0.6421875320374966,0.9990740418434143],"occluded":false},{"timestamp":23.14,"box":[0.2723958492279053,0.287962943315506,0.6661458611488342,0.9564814269542694],"occluded":false},{"timestamp":23.400000000000002,"box":[0.5276042222976685,0.3083333373069763,0.84270840883255,0.9537037014961243],"occluded":false},{"timestamp":23.66,"box":[0.7473958730697632,0.41111111640930176,0.9994792342185974,0.9712963104248047],"occluded":false}],"mean_score":0.8827838897705078},{"noun_phrase":"stone trough","object_id":3,"presentation_id":5,"observations":[{"timestamp":19.5,"box":[0.19947917759418488,0.032407406717538834,0.9994792491197586,0.9694443978369236],"occluded":false},{"timestamp":19.76,"box":[0.2005208432674408,0.032407406717538834,0.999479204416275,0.9703703559935093],"occluded":false},{"timestamp":20.02,"box":[0.20208334922790527,0.032407406717538834,0.9890625476837158,0.9703703559935093],"occluded":false},{"timestamp":20.28,"box":[0.2005208432674408,0.032407406717538834,0.9359375536441803,0.9703703559935093],"occluded":false},{"timestamp":20.54,"box":[0.18802084028720856,0.032407406717538834,0.960937574505806,0.9703703559935093],"occluded":false},{"timestamp":20.8,"box":[0.17760416865348816,0.031481482088565826,0.9494791924953461,0.9703703299164772],"occluded":false},{"timestamp":21.06,"box":[0.14739583432674408,0.031481482088565826,0.9994792193174362,0.9703703299164772],"occluded":false},{"timestamp":21.32,"box":[0.0807291716337204,0.032407406717538834,0.9354167431592941,0.9703703559935093],"occluded":false},{"timestamp":21.580000000000002,"box":[0.0,0.031481482088565826,0.9427083730697632,0.9712962880730629],"occluded":false}],"mean_score":0.6878048777580261},{"noun_phrase":"stone trough","object_id":5,"presentation_id":6,"observations":[{"timestamp":24.18,"box":[0.03958333656191826,0.03333333134651184,0.9994792453944683,0.9694444239139557],"occluded":false},{"timestamp":24.44,"box":[0.10208334028720856,0.032407406717538834,0.9994792491197586,0.9703703559935093],"occluded":false},{"timestamp":24.7,"box":[0.13437500596046448,0.031481482088565826,0.999479204416275,0.9703703299164772],"occluded":false},{"timestamp":24.96,"box":[0.15000000596046448,0.031481482088565826,0.999479204416275,0.9712962880730629],"occluded":false},{"timestamp":25.22,"box":[0.09479167312383652,0.031481482088565826,0.9994791969656944,0.9703703299164772],"occluded":false},{"timestamp":25.48,"box":[0.02239583432674408,0.031481482088565826,0.9994792193174362,0.9712962880730629],"occluded":false},{"timestamp":25.740000000000002,"box":[0.0,0.031481482088565826,0.9994792342185974,0.9703703299164772],"occluded":false},{"timestamp":26.0,"box":[0.0,0.03055555559694767,0.9994792342185974,0.9712962601333857],"occluded":false},{"timestamp":26.26,"box":[0.004687500186264515,0.03055555559694767,0.9994792463257909,0.9703703615814447],"occluded":false},{"timestamp":26.52,"box":[0.05572916939854622,0.031481482088565826,0.9994791932404041,0.9712962880730629],"occluded":false},{"timestamp":26.78,"box":[0.12135417014360428,0.031481482088565826,0.999479241669178,0.9712962880730629],"occluded":false},{"timestamp":27.04,"box":[0.1484375149011612,0.031481482088565826,0.9994792491197586,0.9712962880730629],"occluded":false},{"timestamp":27.3,"box":[0.07395833730697632,0.032407406717538834,0.9994792342185974,0.9712962545454502],"occluded":false},{"timestamp":27.560000000000002,"box":[0.18125000596046448,0.03333333134651184,0.999479204416275,0.9712962806224823],"occluded":false},{"timestamp":27.82,"box":[0.21562501788139343,0.07685185223817825,0.999479204416275,0.9712962582707405],"occluded":false},{"timestamp":28.080000000000002,"box":[0.2447916865348816,0.1509259194135666,0.9994792342185974,0.9722221940755844],"occluded":false},{"timestamp":28.34,"box":[0.24010418355464935,0.20555555820465088,0.9994792193174362,0.9712963104248047],"occluded":false},{"timestamp":28.6,"box":[0.22708334028720856,0.24722221493721008,0.9994791895151138,0.9712962806224823],"occluded":false},{"timestamp":28.86,"box":[0.20208334922790527,0.27129629254341125,0.9994792342185974,0.9712962806224823],"occluded":false},{"timestamp":29.12,"box":[0.18854168057441711,0.277777761220932,0.999479204416275,0.9712962806224823],"occluded":false},{"timestamp":29.38,"box":[0.33802086114883423,0.2787036895751953,0.9994792342185974,0.9694443941116333],"occluded":false}],"mean_score":0.7815700173377991},{"noun_phrase":"stone trough","object_id":2,"presentation_id":7,"observations":[{"timestamp":5.98,"box":[0.33229169249534607,0.6175925731658936,0.46145837008953094,0.8370370119810104],"occluded":false},{"timestamp":6.24,"box":[0.27812501788139343,0.625,0.4177083522081375,0.8490740656852722],"occluded":false},{"timestamp":6.5,"box":[0.2640625238418579,0.6370370388031006,0.4088542014360428,0.8740740716457367],"occluded":false},{"timestamp":6.76,"box":[0.2875000238418579,0.5814814567565918,0.43645836412906647,0.8212962746620178],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.34583336114883423,0.5694444179534912,0.49635420739650726,0.8148147910833359],"occluded":false},{"timestamp":7.28,"box":[0.3661458492279053,0.54537034034729,0.5239583551883698,0.8046295940876007],"occluded":false},{"timestamp":7.54,"box":[0.35572919249534607,0.5333333015441895,0.5244791954755783,0.8240740299224854],"occluded":false},{"timestamp":7.8,"box":[0.38229167461395264,0.5462962985038757,0.5666666775941849,0.8731481432914734],"occluded":false},{"timestamp":8.06,"box":[0.4333333671092987,0.5,0.6348958760499954,0.8601851761341095],"occluded":false},{"timestamp":8.32,"box":[0.4229166805744171,0.5990740656852722,0.6609375327825546,0.9712962806224823],"occluded":false},{"timestamp":8.58,"box":[0.4098958671092987,0.5638889074325562,0.6677083671092987,0.9990740716457367],"occluded":false},{"timestamp":8.84,"box":[0.35989585518836975,0.5074074268341064,0.6458333730697632,0.9712963104248047],"occluded":false},{"timestamp":9.1,"box":[0.34687501192092896,0.49074074625968933,0.6765625178813934,0.9712962806224823],"occluded":false},{"timestamp":9.36,"box":[0.37135419249534607,0.4203703701496124,0.7484375536441803,0.9712962806224823],"occluded":false},{"timestamp":9.620000000000001,"box":[0.38072919845581055,0.42499998211860657,0.9218750596046448,0.9712962806224823],"occluded":false},{"timestamp":9.88,"box":[0.3453125059604645,0.23888888955116272,0.8947916924953461,0.9712962806224823],"occluded":false},{"timestamp":10.14,"box":[0.2770833373069763,0.15555554628372192,0.8296875357627869,0.9712962508201599],"occluded":false},{"timestamp":10.4,"box":[0.2458333522081375,0.12314814329147339,0.9796875864267349,0.9712962508201599],"occluded":false},{"timestamp":10.66,"box":[0.24947917461395264,0.031481482088565826,0.9250000715255737,0.9712962880730629],"occluded":false},{"timestamp":10.92,"box":[0.2421875149011612,0.031481482088565826,0.9489584118127823,0.9703703299164772],"occluded":false},{"timestamp":11.18,"box":[0.24687501788139343,0.031481482088565826,0.9718750417232513,0.9703703299164772],"occluded":false},{"timestamp":11.44,"box":[0.2536458373069763,0.031481482088565826,0.9781250357627869,0.9703703299164772],"occluded":false},{"timestamp":11.700000000000001,"box":[0.26770836114883423,0.031481482088565826,0.9994792342185974,0.9712962880730629],"occluded":false},{"timestamp":11.96,"box":[0.27916666865348816,0.07777777314186096,0.999479204416275,0.9712962806224823],"occluded":false},{"timestamp":12.22,"box":[0.2822916805744171,0.17222221195697784,0.9911458790302277,0.9712962657213211],"occluded":false},{"timestamp":12.48,"box":[0.28177085518836975,0.21574074029922485,0.9828125536441803,0.9712963104248047],"occluded":false},{"timestamp":12.74,"box":[0.28125,0.18981480598449707,0.979687511920929,0.970370352268219],"occluded":false},{"timestamp":13.0,"box":[0.2744791805744171,0.15648147463798523,0.9859375655651093,0.9712962806224823],"occluded":false},{"timestamp":13.26,"box":[0.26875001192092896,0.11203703284263611,0.9875000715255737,0.9712962806224823],"occluded":false},{"timestamp":13.52,"box":[0.25989586114883423,0.032407406717538834,0.9906250834465027,0.9712962545454502],"occluded":false},{"timestamp":13.780000000000001,"box":[0.24062500894069672,0.031481482088565826,0.9718750566244125,0.9703703299164772],"occluded":false},{"timestamp":14.040000000000001,"box":[0.2265625149011612,0.031481482088565826,0.9479167014360428,0.9703703299164772],"occluded":false},{"timestamp":14.3,"box":[0.22343751788139343,0.031481482088565826,0.9468750655651093,0.9712962880730629],"occluded":false},{"timestamp":14.56,"box":[0.22500000894069672,0.031481482088565826,0.9520833939313889,0.9703703299164772],"occluded":false},{"timestamp":14.82,"box":[0.23645834624767303,0.031481482088565826,0.9651042073965073,0.9703703299164772],"occluded":false},{"timestamp":15.08,"box":[0.24270834028720856,0.031481482088565826,0.9713542014360428,0.9703703299164772],"occluded":false},{"timestamp":15.34,"box":[0.24427084624767303,0.031481482088565826,0.9671875685453415,0.9703703299164772],"occluded":false},{"timestamp":15.6,"box":[0.21302084624767303,0.031481482088565826,0.9244792312383652,0.9703703299164772],"occluded":false},{"timestamp":15.860000000000001,"box":[0.04635417088866234,0.06574074178934097,0.7901042066514492,0.9712962880730629],"occluded":false},{"timestamp":16.12,"box":[0.0,0.21296295523643494,0.7031250596046448,0.9712962806224823],"occluded":false},{"timestamp":16.38,"box":[0.0,0.28148147463798523,0.6697916984558105,0.9712962806224823],"occluded":false},{"timestamp":16.64,"box":[0.22708334028720856,0.5935184955596924,0.6656250208616257,0.9712962806224823],"occluded":false},{"timestamp":16.9,"box":[0.3697916865348816,0.5870370268821716,0.6708333790302277,0.9712962806224823],"occluded":false},{"timestamp":17.16,"box":[0.31041666865348816,0.5731481313705444,0.6953125298023224,0.9712962806224823],"occluded":false},{"timestamp":17.42,"box":[0.2682291865348816,0.49259257316589355,0.7588542103767395,0.9712962806224823],"occluded":false},{"timestamp":17.68,"box":[0.11510417610406876,0.08518518507480621,0.835937537252903,0.9712962657213211],"occluded":false},{"timestamp":17.94,"box":[0.17916667461395264,0.032407406717538834,0.9994792342185974,0.9712962545454502],"occluded":false},{"timestamp":18.2,"box":[0.19843751192092896,0.032407406717538834,0.9703125357627869,0.9712962545454502],"occluded":false},{"timestamp":18.46,"box":[0.20260417461395264,0.032407406717538834,0.9994792342185974,0.9712962545454502],"occluded":false},{"timestamp":18.72,"box":[0.2005208432674408,0.032407406717538834,0.999479204416275,0.9712962545454502],"occluded":false}],"mean_score":0.8202247023582458}],"dropped":{},"fates":[{"noun_phrase":"man with dreadlocks","object_id":0,"frames":53,"mean_score":0.6831682920455933,"mean_area":0.14844735182921204,"outcome":"kept","presentation_id":2,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"stone trough","object_id":0,"frames":85,"mean_score":0.6521739363670349,"mean_area":0.22208194005943963,"outcome":"kept","presentation_id":1,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"stone trough","object_id":1,"frames":10,"mean_score":0.7142857313156128,"mean_area":0.0208108133414421,"outcome":"kept","presentation_id":3,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"stone trough","object_id":2,"frames":50,"mean_score":0.8202247023582458,"mean_area":0.43574246970878056,"outcome":"kept","presentation_id":7,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"stone trough","object_id":3,"frames":9,"mean_score":0.6878048777580261,"mean_area":0.7626993575888394,"outcome":"kept","presentation_id":5,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"stone trough","object_id":4,"frames":7,"mean_score":0.8827838897705078,"mean_area":0.28211104058931574,"outcome":"kept","presentation_id":4,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"stone trough","object_id":5,"frames":21,"mean_score":0.7815700173377991,"mean_area":0.7676083258727336,"outcome":"kept","presentation_id":6,"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/tmp/rmot_smoke_real_86260/renders/--cif7Nv6BI_chunk_011_scene_001_choices.mp4","media_bytes":12578833,"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","temperature":0.0,"as_video":true,"media_bytes":12578833,"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 man with dreadlocks who walks towards the stone building\n2. the stone trough filled with water\n3. the dry stone trough in the foreground","questions":["the man with dreadlocks who walks towards the stone building","the stone trough filled with water","the dry stone trough in the foreground"]},"outputs":{"raw_response":"{\n  \"answers\": [\n    {\n      \"description_index\": 1,\n      \"matching_ids\": [\n        2\n      ],\n      \"start_frame\": 2,\n      \"end_frame\": 22,\n      \"confidence\": 1.0\n    },\n    {\n      \"description_index\": 2,\n      \"matching_ids\": [\n        1\n      ],\n      \"start_frame\": 7,\n      \"end_frame\": 30,\n      \"confidence\": 1.0\n    },\n    {\n      \"description_index\": 3,\n      \"matching_ids\": [\n        6,\n        7\n      ],\n      \"start_frame\": 7,\n      \"end_frame\": 30,\n      \"confidence\": 1.0\n    }\n  ]\n}","raw_answers":[{"description_index":1,"matching_ids":[2],"start_frame":2,"end_frame":22,"confidence":1.0},{"description_index":2,"matching_ids":[1],"start_frame":7,"end_frame":30,"confidence":1.0},{"description_index":3,"matching_ids":[6,7],"start_frame":7,"end_frame":30,"confidence":1.0}],"answers":[{"expression":"the man with dreadlocks who walks towards the stone building","matching_presentation_ids":[2],"start_choice_frame":2,"end_choice_frame":22,"confidence":1.0,"self_reported_confidence":1.0},{"expression":"the stone trough filled with water","matching_presentation_ids":[1],"start_choice_frame":7,"end_choice_frame":30,"confidence":1.0,"self_reported_confidence":1.0},{"expression":"the dry stone trough in the foreground","matching_presentation_ids":[6,7],"start_choice_frame":7,"end_choice_frame":30,"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 man with dreadlocks who walks towards the stone building","kind":"motion","is_negative":false,"tracks":[{"timestamps":[1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,13.0,13.26,13.52,13.780000000000001,14.040000000000001,14.3,14.56,14.82,15.08,15.34,15.6,15.860000000000001,17.42,17.68,17.94,18.2,18.46,18.72,18.98,19.240000000000002,19.5,19.76,20.02,20.28,20.54,20.8,21.06,21.32],"boxes":[[0.5625,0.5129629373550415,0.6130208373069763,0.8305555284023285],[0.5291666984558105,0.4981481432914734,0.5802083685994148,0.8296296298503876],[0.5453125238418579,0.4851851761341095,0.6005208604037762,0.8388888835906982],[0.5385416746139526,0.49444442987442017,0.5979166775941849,0.8657407164573669],[0.48593753576278687,0.47777777910232544,0.5458333715796471,0.864814817905426],[0.47083336114883423,0.470370352268219,0.5375000312924385,0.8907407224178314],[0.4645833671092987,0.4749999940395355,0.539062537252903,0.9342592358589172],[0.4593750238418579,0.4120370149612427,0.543750025331974,0.8999999761581421],[0.43281251192092896,0.43611109256744385,0.5296875163912773,0.9712962508201599],[0.40156251192092896,0.4120370149612427,0.4984375163912773,0.9694443941116333],[0.40833336114883423,0.42407405376434326,0.49531253427267075,0.9722222089767456],[0.4411458671092987,0.3574073910713196,0.5338542088866234,0.9731481075286865],[0.4177083671092987,0.35555553436279297,0.5286458730697632,0.9712962508201599],[0.3880208432674408,0.29629629850387573,0.526562511920929,0.9722222089767456],[0.3421875238418579,0.22499999403953552,0.5364583730697632,0.9712962806224823],[0.26354166865348816,0.04537037014961243,0.5432291924953461,0.9712962806224823],[0.10104167461395264,0.032407406717538834,0.4921875298023224,0.9722222127020359],[0.0,0.028703702613711357,0.47447919845581055,0.9749999772757292],[0.0,0.028703702613711357,0.4401041865348816,0.9731481205672026],[0.0,0.00277777761220932,0.3333333432674408,0.9722222313284874],[0.0,0.0,0.29895836114883423,0.9722222089767456],[0.0,0.03055555559694767,0.2838541865348816,0.9722222182899714],[0.0,0.0416666641831398,0.19114583730697632,0.9722221866250038],[0.0,0.4194444417953491,0.2630208432674408,0.9712963104248047],[0.0,0.11388888955116272,0.2682291865348816,0.9712962806224823],[0.0,0.028703702613711357,0.2640625238418579,0.8990740310400724],[0.0,0.02777777798473835,0.24687501788139343,0.7861111033707857],[0.0,0.02777777798473835,0.22812500596046448,0.749074088409543],[0.0,0.02777777798473835,0.2265625149011612,0.7916666734963655],[0.0,0.02777777798473835,0.22552084922790527,0.7861111033707857],[0.0,0.02777777798473835,0.23750001192092896,0.7907407153397799],[0.0,0.026851851493120193,0.24375000596046448,0.7953703589737415],[0.0,0.02777777798473835,0.24687501788139343,0.8259259331971407],[0.0,0.028703702613711357,0.2161458432674408,0.8925925623625517],[0.0,0.635185182094574,0.06979166716337204,0.9638888835906982],[0.0,0.8685185313224792,0.013541667722165585,0.9722222313284874],[0.0,0.5157407522201538,0.11614584177732468,0.972222238779068],[0.0,0.1675925850868225,0.18541668355464935,0.950925886631012],[0.0,0.046296294778585434,0.20208334922790527,0.9314814768731594],[0.0,0.03055555559694767,0.2109375149011612,0.9231481049209833],[0.0,0.03055555559694767,0.20781251788139343,0.9212963078171015],[0.0,0.032407406717538834,0.20677083730697632,0.941666666418314],[0.0,0.031481482088565826,0.20781251788139343,0.9370370283722878],[0.0,0.029629629105329514,0.20677083730697632,0.9351851753890514],[0.0,0.029629629105329514,0.20781251788139343,0.9490740709006786],[0.0,0.03055555559694767,0.2083333432674408,0.9601851794868708],[0.0,0.029629629105329514,0.2057291716337204,0.9638888649642467],[0.0,0.028703702613711357,0.26875001192092896,0.9324073921889067],[0.0,0.029629629105329514,0.18958334624767303,0.8888888768851757],[0.0,0.10462962836027145,0.21041667461395264,0.8592592403292656],[0.0,0.520370364189148,0.10000000894069672,0.9314814805984497]]}]},{"expression":"the stone trough filled with water","kind":"appearance","is_negative":false,"tracks":[{"timestamps":[5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88,10.14,10.4,10.66,10.92,11.18,11.44,11.700000000000001,11.96,12.22,12.48,12.74,13.0,13.26,13.52,13.780000000000001,14.040000000000001,14.3,14.56,14.82,15.08,15.34,15.6,15.860000000000001,16.12,16.38,16.64,16.9,17.16,17.42,17.68,17.94,18.2,18.46,18.72,18.98,19.240000000000002,19.5,19.76,20.02,20.28,20.54,20.8,21.06,21.32,21.580000000000002,21.84,22.1,23.400000000000002,23.66,23.92,24.18,24.44,24.7,24.96,25.22,26.52,26.78,27.04,27.3,27.560000000000002,27.82,28.080000000000002,28.34,28.6,28.86,29.12,29.38],"boxes":[[0.4270833432674408,0.6064814925193787,0.5593750178813934,0.7962962985038757],[0.3812500238418579,0.5944444537162781,0.520833358168602,0.7962962985038757],[0.3671875298023224,0.5999999642372131,0.5109375417232513,0.8175925463438034],[0.39479169249534607,0.5546296238899231,0.546354204416275,0.7824074029922485],[0.4515625238418579,0.539814829826355,0.6057291924953461,0.7777777910232544],[0.4776041805744171,0.5148147940635681,0.6432291865348816,0.7648147940635681],[0.4713541865348816,0.5046296119689941,0.6447916924953461,0.7842592298984528],[0.49531251192092896,0.5231481194496155,0.6942708492279053,0.8287036716938019],[0.5479167103767395,0.4851851761341095,0.778125062584877,0.8296296000480652],[0.5442708730697632,0.5777777433395386,0.8135417103767395,0.9666666388511658],[0.5401042103767395,0.5407407283782959,0.8364583849906921,0.9685184955596924],[0.4958333671092987,0.4749999940395355,0.8083333969116211,0.9361110925674438],[0.4932291805744171,0.46759259700775146,0.8479166924953461,0.9722222089767456],[0.5213541984558105,0.39629629254341125,0.9447917342185974,0.9703703820705414],[0.5458333492279053,0.3870370388031006,0.999479204416275,0.9712963104248047],[0.5390625,0.20462962985038757,0.999479204416275,0.9675925672054291],[0.5020833611488342,0.10925925523042679,0.9994792342185974,0.9222222045063972],[0.49531251192092896,0.07407407462596893,0.9994792342185974,0.963888868689537],[0.5161458849906921,0.031481482088565826,0.9994792342185974,0.8601851835846901],[0.5411458611488342,0.029629629105329514,0.999479204416275,0.7731481604278088],[0.5494791865348816,0.03055555559694767,0.999479204416275,0.7509259078651667],[0.5546875,0.03055555559694767,0.999479204416275,0.7999999616295099],[0.5614583492279053,0.031481482088565826,0.999479204416275,0.8518518581986427],[0.5739583373069763,0.031481482088565826,0.999479204416275,0.871296264231205],[0.5802083611488342,0.060185182839632034,0.999479204416275,0.9231481440365314],[0.5817708373069763,0.1527777761220932,0.999479204416275,0.9685184806585312],[0.581250011920929,0.13055555522441864,0.999479204416275,0.963888868689537],[0.5770833492279053,0.0962962955236435,0.999479204416275,0.9370370358228683],[0.5776041746139526,0.054629627615213394,0.999479204416275,0.9083333052694798],[0.5755208730697632,0.031481482088565826,0.9994792342185974,0.8749999776482582],[0.5703125,0.03055555559694767,0.9994791746139526,0.7555555198341608],[0.550000011920929,0.03055555559694767,0.999479204416275,0.7768518421798944],[0.5520833730697632,0.03055555559694767,0.9994792342185974,0.7879629228264093],[0.5651041865348816,0.03055555559694767,0.999479204416275,0.8249999973922968],[0.5708333849906921,0.03055555559694767,0.9994792342185974,0.8296296093612909],[0.5713542103767395,0.03055555559694767,0.9994792342185974,0.8379629347473383],[0.565625011920929,0.03055555559694767,0.999479204416275,0.8407407496124506],[0.5281250476837158,0.031481482088565826,0.9994792342185974,0.8148147836327553],[0.12968750298023224,0.032407406717538834,0.9994791895151138,0.9527777470648289],[0.30677086114883423,0.11296296119689941,0.9041666984558105,0.8879629373550415],[0.28125,0.1527777761220932,0.8640625476837158,0.963888868689537],[0.26770836114883423,0.18611110746860504,0.8489583730697632,0.9685185104608536],[0.2734375,0.182407408952713,0.8562500476837158,0.9703703820705414],[0.2968750298023224,0.15555554628372192,0.8911459147930145,0.9657407402992249],[0.3552083373069763,0.0824074074625969,0.9942708611488342,0.9148148223757744],[0.4307291805744171,0.03055555559694767,0.999479204416275,0.914814779534936],[0.48750001192092896,0.031481482088565826,0.9994792342185974,0.8870370164513588],[0.5088542103767395,0.031481482088565826,0.9994792342185974,0.8518518581986427],[0.5145833492279053,0.031481482088565826,0.999479204416275,0.8870370164513588],[0.5104166865348816,0.03055555559694767,0.999479204416275,0.8592592570930719],[0.5437500476837158,0.03055555559694767,0.9994792342185974,0.40462962724268436],[0.6296875476837158,0.031481482088565826,0.9994792342185974,0.40833333879709244],[0.6338542103767395,0.03055555559694767,0.9994792342185974,0.4092592392116785],[0.6234375238418579,0.03055555559694767,0.999479204416275,0.4222222063690424],[0.6182292103767395,0.031481482088565826,0.9994792342185974,0.43611110001802444],[0.6083333492279053,0.031481482088565826,0.999479204416275,0.4324073866009712],[0.6270833611488342,0.03055555559694767,0.999479204416275,0.4074074123054743],[0.6640625596046448,0.03055555559694767,0.9994792342185974,0.3379629645496607],[0.6364583373069763,0.03055555559694767,0.999479204416275,0.33333332277834415],[0.518750011920929,0.03055555559694767,0.999479204416275,0.5259259138256311],[0.48750001192092896,0.03055555559694767,0.8401041924953461,0.22685184516012669],[0.15416666865348816,0.031481482088565826,0.5604166984558105,0.20185185223817825],[0.0,0.12314814329147339,0.17239584028720856,0.3083333224058151],[0.0,0.17129629850387573,0.2588541805744171,0.364814817905426],[0.0,0.046296294778585434,0.550000011920929,0.28333332762122154],[0.12656250596046448,0.03055555559694767,0.8140625655651093,0.27500000037252903],[0.5234375,0.031481482088565826,0.8843750059604645,0.31111110001802444],[0.581250011920929,0.031481482088565826,0.8411458730697632,0.2749999985098839],[0.6822916865348816,0.031481482088565826,0.8963541984558105,0.23888888210058212],[0.745312511920929,0.03055555559694767,0.9244791865348816,0.15000000037252903],[0.7604166865348816,0.031481482088565826,0.8942708671092987,0.10277777910232544],[0.7541667222976685,0.03055555559694767,0.8734375610947609,0.11296296305954456],[0.7307292222976685,0.03055555559694767,0.9572917371988297,0.24074074067175388],[0.7015625238418579,0.031481482088565826,0.9848958849906921,0.32407406717538834],[0.6098958849906921,0.031481482088565826,0.9458334147930145,0.40740741044282913],[0.5494791865348816,0.05648148059844971,0.9177083671092987,0.44722220301628113],[0.5098958611488342,0.11388888955116272,0.8604167103767395,0.4333333373069763],[0.4989583492279053,0.18796296417713165,0.835416704416275,0.45277778804302216],[0.4911458492279053,0.23888888955116272,0.8187500238418579,0.510185182094574],[0.47812503576278687,0.28333333134651184,0.8338542282581329,0.5518518388271332],[0.45520836114883423,0.3046296238899231,0.8395833671092987,0.5833333134651184],[0.4505208432674408,0.3092592656612396,0.8515625298023224,0.5972222089767456],[0.4046875238418579,0.307407408952713,0.8098958730697632,0.5435185134410858]]}]}],"rejected":{"the dry stone trough in the foreground":"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 man with dreadlocks who walks towards the stone building","kind":"motion","is_negative":false,"tracks":[{"timestamps":[1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86,3.12,3.38,3.64,3.9,4.16,4.42,4.68,4.94,5.2,5.46,5.72,5.98,6.24,6.5,6.76,13.0,13.26,13.52,13.780000000000001,14.040000000000001,14.3,14.56,14.82,15.08,15.34,15.6,15.860000000000001,17.42,17.68,17.94,18.2,18.46,18.72,18.98,19.240000000000002,19.5,19.76,20.02,20.28,20.54,20.8,21.06,21.32],"boxes":[[0.5625,0.5129629373550415,0.6130208373069763,0.8305555284023285],[0.5291666984558105,0.4981481432914734,0.5802083685994148,0.8296296298503876],[0.5453125238418579,0.4851851761341095,0.6005208604037762,0.8388888835906982],[0.5385416746139526,0.49444442987442017,0.5979166775941849,0.8657407164573669],[0.48593753576278687,0.47777777910232544,0.5458333715796471,0.864814817905426],[0.47083336114883423,0.470370352268219,0.5375000312924385,0.8907407224178314],[0.4645833671092987,0.4749999940395355,0.539062537252903,0.9342592358589172],[0.4593750238418579,0.4120370149612427,0.543750025331974,0.8999999761581421],[0.43281251192092896,0.43611109256744385,0.5296875163912773,0.9712962508201599],[0.40156251192092896,0.4120370149612427,0.4984375163912773,0.9694443941116333],[0.40833336114883423,0.42407405376434326,0.49531253427267075,0.9722222089767456],[0.4411458671092987,0.3574073910713196,0.5338542088866234,0.9731481075286865],[0.4177083671092987,0.35555553436279297,0.5286458730697632,0.9712962508201599],[0.3880208432674408,0.29629629850387573,0.526562511920929,0.9722222089767456],[0.3421875238418579,0.22499999403953552,0.5364583730697632,0.9712962806224823],[0.26354166865348816,0.04537037014961243,0.5432291924953461,0.9712962806224823],[0.10104167461395264,0.032407406717538834,0.4921875298023224,0.9722222127020359],[0.0,0.028703702613711357,0.47447919845581055,0.9749999772757292],[0.0,0.028703702613711357,0.4401041865348816,0.9731481205672026],[0.0,0.00277777761220932,0.3333333432674408,0.9722222313284874],[0.0,0.0,0.29895836114883423,0.9722222089767456],[0.0,0.03055555559694767,0.2838541865348816,0.9722222182899714],[0.0,0.0416666641831398,0.19114583730697632,0.9722221866250038],[0.0,0.4194444417953491,0.2630208432674408,0.9712963104248047],[0.0,0.11388888955116272,0.2682291865348816,0.9712962806224823],[0.0,0.028703702613711357,0.2640625238418579,0.8990740310400724],[0.0,0.02777777798473835,0.24687501788139343,0.7861111033707857],[0.0,0.02777777798473835,0.22812500596046448,0.749074088409543],[0.0,0.02777777798473835,0.2265625149011612,0.7916666734963655],[0.0,0.02777777798473835,0.22552084922790527,0.7861111033707857],[0.0,0.02777777798473835,0.23750001192092896,0.7907407153397799],[0.0,0.026851851493120193,0.24375000596046448,0.7953703589737415],[0.0,0.02777777798473835,0.24687501788139343,0.8259259331971407],[0.0,0.028703702613711357,0.2161458432674408,0.8925925623625517],[0.0,0.635185182094574,0.06979166716337204,0.9638888835906982],[0.0,0.8685185313224792,0.013541667722165585,0.9722222313284874],[0.0,0.5157407522201538,0.11614584177732468,0.972222238779068],[0.0,0.1675925850868225,0.18541668355464935,0.950925886631012],[0.0,0.046296294778585434,0.20208334922790527,0.9314814768731594],[0.0,0.03055555559694767,0.2109375149011612,0.9231481049209833],[0.0,0.03055555559694767,0.20781251788139343,0.9212963078171015],[0.0,0.032407406717538834,0.20677083730697632,0.941666666418314],[0.0,0.031481482088565826,0.20781251788139343,0.9370370283722878],[0.0,0.029629629105329514,0.20677083730697632,0.9351851753890514],[0.0,0.029629629105329514,0.20781251788139343,0.9490740709006786],[0.0,0.03055555559694767,0.2083333432674408,0.9601851794868708],[0.0,0.029629629105329514,0.2057291716337204,0.9638888649642467],[0.0,0.028703702613711357,0.26875001192092896,0.9324073921889067],[0.0,0.029629629105329514,0.18958334624767303,0.8888888768851757],[0.0,0.10462962836027145,0.21041667461395264,0.8592592403292656],[0.0,0.520370364189148,0.10000000894069672,0.9314814805984497]]}]},{"expression":"the stone trough filled with water","kind":"appearance","is_negative":false,"tracks":[{"timestamps":[5.98,6.24,6.5,6.76,7.0200000000000005,7.28,7.54,7.8,8.06,8.32,8.58,8.84,9.1,9.36,9.620000000000001,9.88,10.14,10.4,10.66,10.92,11.18,11.44,11.700000000000001,11.96,12.22,12.48,12.74,13.0,13.26,13.52,13.780000000000001,14.040000000000001,14.3,14.56,14.82,15.08,15.34,15.6,15.860000000000001,16.12,16.38,16.64,16.9,17.16,17.42,17.68,17.94,18.2,18.46,18.72,18.98,19.240000000000002,19.5,19.76,20.02,20.28,20.54,20.8,21.06,21.32,21.580000000000002,21.84,22.1,23.400000000000002,23.66,23.92,24.18,24.44,24.7,24.96,25.22,26.52,26.78,27.04,27.3,27.560000000000002,27.82,28.080000000000002,28.34,28.6,28.86,29.12,29.38],"boxes":[[0.4270833432674408,0.6064814925193787,0.5593750178813934,0.7962962985038757],[0.3812500238418579,0.5944444537162781,0.520833358168602,0.7962962985038757],[0.3671875298023224,0.5999999642372131,0.5109375417232513,0.8175925463438034],[0.39479169249534607,0.5546296238899231,0.546354204416275,0.7824074029922485],[0.4515625238418579,0.539814829826355,0.6057291924953461,0.7777777910232544],[0.4776041805744171,0.5148147940635681,0.6432291865348816,0.7648147940635681],[0.4713541865348816,0.5046296119689941,0.6447916924953461,0.7842592298984528],[0.49531251192092896,0.5231481194496155,0.6942708492279053,0.8287036716938019],[0.5479167103767395,0.4851851761341095,0.778125062584877,0.8296296000480652],[0.5442708730697632,0.5777777433395386,0.8135417103767395,0.9666666388511658],[0.5401042103767395,0.5407407283782959,0.8364583849906921,0.9685184955596924],[0.4958333671092987,0.4749999940395355,0.8083333969116211,0.9361110925674438],[0.4932291805744171,0.46759259700775146,0.8479166924953461,0.9722222089767456],[0.5213541984558105,0.39629629254341125,0.9447917342185974,0.9703703820705414],[0.5458333492279053,0.3870370388031006,0.999479204416275,0.9712963104248047],[0.5390625,0.20462962985038757,0.999479204416275,0.9675925672054291],[0.5020833611488342,0.10925925523042679,0.9994792342185974,0.9222222045063972],[0.49531251192092896,0.07407407462596893,0.9994792342185974,0.963888868689537],[0.5161458849906921,0.031481482088565826,0.9994792342185974,0.8601851835846901],[0.5411458611488342,0.029629629105329514,0.999479204416275,0.7731481604278088],[0.5494791865348816,0.03055555559694767,0.999479204416275,0.7509259078651667],[0.5546875,0.03055555559694767,0.999479204416275,0.7999999616295099],[0.5614583492279053,0.031481482088565826,0.999479204416275,0.8518518581986427],[0.5739583373069763,0.031481482088565826,0.999479204416275,0.871296264231205],[0.5802083611488342,0.060185182839632034,0.999479204416275,0.9231481440365314],[0.5817708373069763,0.1527777761220932,0.999479204416275,0.9685184806585312],[0.581250011920929,0.13055555522441864,0.999479204416275,0.963888868689537],[0.5770833492279053,0.0962962955236435,0.999479204416275,0.9370370358228683],[0.5776041746139526,0.054629627615213394,0.999479204416275,0.9083333052694798],[0.5755208730697632,0.031481482088565826,0.9994792342185974,0.8749999776482582],[0.5703125,0.03055555559694767,0.9994791746139526,0.7555555198341608],[0.550000011920929,0.03055555559694767,0.999479204416275,0.7768518421798944],[0.5520833730697632,0.03055555559694767,0.9994792342185974,0.7879629228264093],[0.5651041865348816,0.03055555559694767,0.999479204416275,0.8249999973922968],[0.5708333849906921,0.03055555559694767,0.9994792342185974,0.8296296093612909],[0.5713542103767395,0.03055555559694767,0.9994792342185974,0.8379629347473383],[0.565625011920929,0.03055555559694767,0.999479204416275,0.8407407496124506],[0.5281250476837158,0.031481482088565826,0.9994792342185974,0.8148147836327553],[0.12968750298023224,0.032407406717538834,0.9994791895151138,0.9527777470648289],[0.30677086114883423,0.11296296119689941,0.9041666984558105,0.8879629373550415],[0.28125,0.1527777761220932,0.8640625476837158,0.963888868689537],[0.26770836114883423,0.18611110746860504,0.8489583730697632,0.9685185104608536],[0.2734375,0.182407408952713,0.8562500476837158,0.9703703820705414],[0.2968750298023224,0.15555554628372192,0.8911459147930145,0.9657407402992249],[0.3552083373069763,0.0824074074625969,0.9942708611488342,0.9148148223757744],[0.4307291805744171,0.03055555559694767,0.999479204416275,0.914814779534936],[0.48750001192092896,0.031481482088565826,0.9994792342185974,0.8870370164513588],[0.5088542103767395,0.031481482088565826,0.9994792342185974,0.8518518581986427],[0.5145833492279053,0.031481482088565826,0.999479204416275,0.8870370164513588],[0.5104166865348816,0.03055555559694767,0.999479204416275,0.8592592570930719],[0.5437500476837158,0.03055555559694767,0.9994792342185974,0.40462962724268436],[0.6296875476837158,0.031481482088565826,0.9994792342185974,0.40833333879709244],[0.6338542103767395,0.03055555559694767,0.9994792342185974,0.4092592392116785],[0.6234375238418579,0.03055555559694767,0.999479204416275,0.4222222063690424],[0.6182292103767395,0.031481482088565826,0.9994792342185974,0.43611110001802444],[0.6083333492279053,0.031481482088565826,0.999479204416275,0.4324073866009712],[0.6270833611488342,0.03055555559694767,0.999479204416275,0.4074074123054743],[0.6640625596046448,0.03055555559694767,0.9994792342185974,0.3379629645496607],[0.6364583373069763,0.03055555559694767,0.999479204416275,0.33333332277834415],[0.518750011920929,0.03055555559694767,0.999479204416275,0.5259259138256311],[0.48750001192092896,0.03055555559694767,0.8401041924953461,0.22685184516012669],[0.15416666865348816,0.031481482088565826,0.5604166984558105,0.20185185223817825],[0.0,0.12314814329147339,0.17239584028720856,0.3083333224058151],[0.0,0.17129629850387573,0.2588541805744171,0.364814817905426],[0.0,0.046296294778585434,0.550000011920929,0.28333332762122154],[0.12656250596046448,0.03055555559694767,0.8140625655651093,0.27500000037252903],[0.5234375,0.031481482088565826,0.8843750059604645,0.31111110001802444],[0.581250011920929,0.031481482088565826,0.8411458730697632,0.2749999985098839],[0.6822916865348816,0.031481482088565826,0.8963541984558105,0.23888888210058212],[0.745312511920929,0.03055555559694767,0.9244791865348816,0.15000000037252903],[0.7604166865348816,0.031481482088565826,0.8942708671092987,0.10277777910232544],[0.7541667222976685,0.03055555559694767,0.8734375610947609,0.11296296305954456],[0.7307292222976685,0.03055555559694767,0.9572917371988297,0.24074074067175388],[0.7015625238418579,0.031481482088565826,0.9848958849906921,0.32407406717538834],[0.6098958849906921,0.031481482088565826,0.9458334147930145,0.40740741044282913],[0.5494791865348816,0.05648148059844971,0.9177083671092987,0.44722220301628113],[0.5098958611488342,0.11388888955116272,0.8604167103767395,0.4333333373069763],[0.4989583492279053,0.18796296417713165,0.835416704416275,0.45277778804302216],[0.4911458492279053,0.23888888955116272,0.8187500238418579,0.510185182094574],[0.47812503576278687,0.28333333134651184,0.8338542282581329,0.5518518388271332],[0.45520836114883423,0.3046296238899231,0.8395833671092987,0.5833333134651184],[0.4505208432674408,0.3092592656612396,0.8515625298023224,0.5972222089767456],[0.4046875238418579,0.307407408952713,0.8098958730697632,0.5435185134410858]]}]}],"rmot_expression_count":2,"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","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 bearded man in a black hoodie who appears on the right\",\n      \"noun_phrase\": \"bearded man in black hoodie\",\n      \"kind\": \"appearance\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the white monument on top of the rocky hill in the distance\",\n      \"noun_phrase\": \"white monument\",\n      \"kind\": \"spatial\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the person walking on the stone wall\",\n      \"noun_phrase\": \"person\",\n      \"kind\": \"motion\",\n      \"expected_count\": 0,\n      \"is_negative\": true\n    }\n  ]\n}","raw_items":[{"expression":"the bearded man in a black hoodie who appears on the right","noun_phrase":"bearded man in black hoodie","kind":"appearance","expected_count":1,"is_negative":false},{"expression":"the white monument on top of the rocky hill in the distance","noun_phrase":"white monument","kind":"spatial","expected_count":1,"is_negative":false},{"expression":"the person walking on the stone wall","noun_phrase":"person","kind":"motion","expected_count":0,"is_negative":true}],"accepted":[{"expression":"the bearded man in a black hoodie who appears on the right","noun_phrase":"bearded man in black hoodie","kind":"appearance","expected_count":1,"is_negative":false},{"expression":"the white monument on top of the rocky hill in the distance","noun_phrase":"white monument","kind":"spatial","expected_count":1,"is_negative":false},{"expression":"the person walking on the stone wall","noun_phrase":"person","kind":"motion","expected_count":0,"is_negative":true}],"gated":[]}},"track":{"inputs":{"tracker":"Sam31MultiplexTracker","model_id":"/mnt/ceph3/ec/jacky/sam3.1","noun_phrases":["bearded man in black hoodie","person","white monument"],"expressions_in":3,"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":3},"outputs":{"raw_tracks":[{"noun_phrase":"bearded man in black hoodie","object_id":0,"observations":[{"timestamp":21.84,"box":[0.8364583849906921,0.40648147463798523,0.9994792342185974,0.9740740358829498],"occluded":false},{"timestamp":22.1,"box":[0.6536458730697632,0.43518516421318054,0.8911458849906921,0.9749999940395355],"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":[0.7064220309257507,0.7064220309257507,0.7064220309257507,0.7064220309257507,0.7064220309257507,0.7064220309257507]},{"noun_phrase":"person","object_id":0,"observations":[{"timestamp":21.84,"box":[0.8348958492279053,0.40648147463798523,0.9994791895151138,0.9740740358829498],"occluded":false},{"timestamp":22.1,"box":[0.6536458730697632,0.43611109256744385,0.8901042193174362,0.9740740656852722],"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":[0.4229166805744171,0.3166666626930237,0.7401041984558105,0.9787036776542664],"occluded":false},{"timestamp":23.14,"box":[0.3916666805744171,0.2537037134170532,0.7036458551883698,0.9759259223937988],"occluded":false}],"scores":[0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708,0.941605806350708]}],"tracklets":[{"noun_phrase":"bearded man in black hoodie","object_id":0,"presentation_id":1,"observations":[{"timestamp":21.84,"box":[0.8364583849906921,0.40648147463798523,0.9994792342185974,0.9740740358829498],"occluded":false},{"timestamp":22.1,"box":[0.6536458730697632,0.43518516421318054,0.8911458849906921,0.9749999940395355],"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":0.7064220309257507}],"dropped":{"merged_duplicate":1},"fates":[{"noun_phrase":"bearded man in black hoodie","object_id":0,"frames":6,"mean_score":0.7064220309257507,"mean_area":0.16548909121392263,"outcome":"kept","presentation_id":1,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"person","object_id":0,"frames":6,"mean_score":0.941605806350708,"mean_area":0.16552766961946208,"outcome":"merged_duplicate","presentation_id":null,"merged_into_phrase":"bearded man in black hoodie","merged_into_object_id":0,"merge_iou":0.9968436982531227}],"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/tmp/rmot_smoke_real_86260/renders/--cif7Nv6BI_chunk_010_scene_016_choices.mp4","media_bytes":7898397,"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","temperature":0.0,"as_video":true,"media_bytes":7898397,"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 bearded man in a black hoodie who appears on the right\n2. the white monument on top of the rocky hill in the distance\n3. the person walking on the stone wall","questions":["the bearded man in a black hoodie who appears on the right","the white monument on top of the rocky hill in the distance","the person walking on 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      \"description_index\": 2,\n      \"matching_ids\": [],\n      \"start_frame\": 1,\n      \"end_frame\": 24,\n      \"confidence\": 1.0\n    },\n    {\n      \"description_index\": 3,\n      \"matching_ids\": [],\n      \"start_frame\": 1,\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},{"description_index":2,"matching_ids":[],"start_frame":1,"end_frame":24,"confidence":1.0},{"description_index":3,"matching_ids":[],"start_frame":1,"end_frame":24,"confidence":1.0}],"answers":[{"expression":"the bearded man in a black hoodie who appears on the right","matching_presentation_ids":[1],"start_choice_frame":23,"end_choice_frame":24,"confidence":1.0,"self_reported_confidence":1.0},{"expression":"the white monument on top of the rocky hill in the distance","matching_presentation_ids":[],"start_choice_frame":1,"end_choice_frame":24,"confidence":1.0,"self_reported_confidence":1.0},{"expression":"the person walking on the stone wall","matching_presentation_ids":[],"start_choice_frame":1,"end_choice_frame":24,"confidence":1.0,"self_reported_confidence":1.0}]}},"verify":{"inputs":{"expressions":3,"tracklets":1,"selections":3,"gates":"interval, count, unknown ids, confidence, then geometry (motion)"},"outputs":{"kept":[{"expression":"the bearded man in a black hoodie who appears on the right","kind":"appearance","is_negative":false,"tracks":[{"timestamps":[22.1,22.36,22.62,22.88,23.14],"boxes":[[0.6536458730697632,0.43518516421318054,0.8911458849906921,0.9749999940395355],[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":{"the person walking on the stone wall":"no_candidates_retrieved","the white monument on top of the rocky hill in the distance":"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_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 bearded man in a black hoodie who appears on the right","kind":"appearance","is_negative":false,"tracks":[{"timestamps":[22.1,22.36,22.62,22.88,23.14],"boxes":[[0.6536458730697632,0.43518516421318054,0.8911458849906921,0.9749999940395355],[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","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 wearing a black cap who is talking to the camera\",\n      \"noun_phrase\": \"man wearing a black cap\",\n      \"kind\": \"interaction\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the man sitting in the back seat\",\n      \"noun_phrase\": \"man\",\n      \"kind\": \"spatial\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the man in the back seat who is not wearing sunglasses\",\n      \"noun_phrase\": \"man\",\n      \"kind\": \"appearance\",\n      \"expected_count\": 0,\n      \"is_negative\": true\n    }\n  ]\n}","raw_items":[{"expression":"the man wearing a black cap who is talking to the camera","noun_phrase":"man wearing a black cap","kind":"interaction","expected_count":1,"is_negative":false},{"expression":"the man sitting in the back seat","noun_phrase":"man","kind":"spatial","expected_count":1,"is_negative":false},{"expression":"the man in the back seat who is not wearing sunglasses","noun_phrase":"man","kind":"appearance","expected_count":0,"is_negative":true}],"accepted":[{"expression":"the man wearing a black cap who is talking to the camera","noun_phrase":"man wearing a black cap","kind":"interaction","expected_count":1,"is_negative":false},{"expression":"the man sitting in the back seat","noun_phrase":"man","kind":"spatial","expected_count":1,"is_negative":false},{"expression":"the man in the back seat who is not wearing sunglasses","noun_phrase":"man","kind":"appearance","expected_count":0,"is_negative":true}],"gated":[]}},"track":{"inputs":{"tracker":"Sam31MultiplexTracker","model_id":"/mnt/ceph3/ec/jacky/sam3.1","noun_phrases":["man","man wearing a black cap"],"expressions_in":3,"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":2},"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]},{"noun_phrase":"man wearing a black cap","object_id":0,"observations":[{"timestamp":0.5338666666666667,"box":[0.0,0.0,0.4453125298023224,0.9990740418434143],"occluded":false},{"timestamp":0.8008000000000001,"box":[0.0,0.0,0.4583333432674408,0.9990740418434143],"occluded":false},{"timestamp":1.0677333333333334,"box":[0.0,0.0,0.9812500476837158,0.9990740418434143],"occluded":false},{"timestamp":1.3346666666666667,"box":[0.0,0.0,0.4828125238418579,0.9990740418434143],"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":[0.0,0.0,0.5218750238418579,0.9990740418434143],"occluded":false},{"timestamp":2.4024,"box":[0.0,0.0,0.47447919845581055,0.9990740418434143],"occluded":false},{"timestamp":2.6693333333333333,"box":[0.0,0.0,0.5208333730697632,0.9990740418434143],"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":[0.0,0.0,0.5614583492279053,0.9990740418434143],"occluded":false},{"timestamp":3.7370666666666668,"box":[0.0,0.0,0.5692708492279053,0.9990740418434143],"occluded":false},{"timestamp":4.0040000000000004,"box":[0.0,0.0,0.5765625238418579,0.9990740418434143],"occluded":false},{"timestamp":4.270933333333334,"box":[0.0,0.0,0.5687500238418579,0.9990740418434143],"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":[0.0,0.0,0.6270833611488342,0.9990740418434143],"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":[0.0,0.0,0.581250011920929,0.9990740418434143],"occluded":false},{"timestamp":7.2072,"box":[0.0,0.0,0.596875011920929,0.9990740418434143],"occluded":false},{"timestamp":7.4741333333333335,"box":[0.0,0.0,0.5973958373069763,0.9990740418434143],"occluded":false},{"timestamp":7.741066666666667,"box":[0.0,0.0,0.5776041746139526,0.9990740418434143],"occluded":false},{"timestamp":8.008000000000001,"box":[0.0,0.0,0.5651041865348816,0.9990740418434143],"occluded":false},{"timestamp":8.274933333333333,"box":[0.0,0.0,0.5875000357627869,0.9990740418434143],"occluded":false},{"timestamp":8.541866666666667,"box":[0.0,0.0,0.5494791865348816,0.9990740418434143],"occluded":false},{"timestamp":8.8088,"box":[0.0,0.0,0.5729166865348816,0.9990740418434143],"occluded":false},{"timestamp":9.075733333333334,"box":[0.0,0.0,0.5526041984558105,0.9990740418434143],"occluded":false},{"timestamp":9.342666666666666,"box":[0.01927083358168602,0.0,0.5744792185723782,0.9990740418434143],"occluded":false},{"timestamp":9.6096,"box":[0.0,0.0,0.5776041746139526,0.9990740418434143],"occluded":false},{"timestamp":9.876533333333334,"box":[0.0,0.0,0.5848958492279053,0.9990740418434143],"occluded":false},{"timestamp":10.143466666666667,"box":[0.0,0.0,0.5729166865348816,0.9990740418434143],"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":[0.0,0.0,0.5895833373069763,0.9990740418434143],"occluded":false},{"timestamp":11.478133333333334,"box":[0.0,0.0,0.6151041984558105,0.9990740418434143],"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":[0.0,0.0,0.6322916746139526,0.9990740418434143],"occluded":false},{"timestamp":12.812800000000001,"box":[0.0,0.0,0.6291667222976685,0.9972221851348877],"occluded":false},{"timestamp":13.079733333333333,"box":[0.0,0.0,0.6234375238418579,0.9990740418434143],"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":[0.0,0.0,0.6296875476837158,0.9990740418434143],"occluded":false},{"timestamp":14.4144,"box":[0.0,0.0,0.6078125238418579,0.9990740418434143],"occluded":false},{"timestamp":14.681333333333335,"box":[0.0,0.0,0.5859375596046448,0.9990740418434143],"occluded":false},{"timestamp":14.948266666666667,"box":[0.0,0.0,0.5859375596046448,0.9990740418434143],"occluded":false},{"timestamp":15.215200000000001,"box":[0.0,0.0,0.5916666984558105,0.9990740418434143],"occluded":false},{"timestamp":15.482133333333334,"box":[0.0,0.0,0.5880208611488342,0.9990740418434143],"occluded":false},{"timestamp":15.749066666666668,"box":[0.0,0.0,0.5833333730697632,0.9990740418434143],"occluded":false},{"timestamp":16.016000000000002,"box":[0.0,0.0,0.5755208730697632,0.9990740418434143],"occluded":false},{"timestamp":16.282933333333332,"box":[0.0,0.0,0.5708333849906921,0.9990740418434143],"occluded":false},{"timestamp":16.549866666666667,"box":[0.0020833334419876337,0.0,0.5437500199768692,0.9990740418434143],"occluded":false}],"scores":[0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258,0.8632478713989258]}],"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":{"merged_duplicate":1},"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},{"noun_phrase":"man wearing a black cap","object_id":0,"frames":61,"mean_score":0.8632478713989258,"mean_area":0.5811618835570428,"outcome":"merged_duplicate","presentation_id":null,"merged_into_phrase":"man","merged_into_object_id":1,"merge_iou":0.9723944257016279}],"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/tmp/rmot_smoke_real_86260/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","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 wearing a black cap who is talking to the camera\n2. the man sitting in the back seat\n3. the man in the back seat who is not wearing sunglasses","questions":["the man wearing a black cap who is talking to the camera","the man sitting in the back seat","the man in the back seat who is not wearing sunglasses"]},"outputs":{"raw_response":"{\n  \"answers\": [\n    {\n      \"description_index\": 1,\n      \"matching_ids\": [\n        1\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      \"description_index\": 3,\n      \"matching_ids\": [],\n      \"start_frame\": 1,\n      \"end_frame\": 18,\n      \"confidence\": 0.9\n    }\n  ]\n}","raw_answers":[{"description_index":1,"matching_ids":[1],"start_frame":1,"end_frame":18,"confidence":1.0},{"description_index":2,"matching_ids":[3],"start_frame":1,"end_frame":18,"confidence":1.0},{"description_index":3,"matching_ids":[],"start_frame":1,"end_frame":18,"confidence":0.9}],"answers":[{"expression":"the man wearing a black cap who is talking to the camera","matching_presentation_ids":[1],"start_choice_frame":1,"end_choice_frame":18,"confidence":1.0,"self_reported_confidence":1.0},{"expression":"the man sitting in the back seat","matching_presentation_ids":[3],"start_choice_frame":1,"end_choice_frame":18,"confidence":1.0,"self_reported_confidence":1.0},{"expression":"the man in the back seat who is not wearing sunglasses","matching_presentation_ids":[],"start_choice_frame":1,"end_choice_frame":18,"confidence":0.9,"self_reported_confidence":0.9}]}},"verify":{"inputs":{"expressions":3,"tracklets":3,"selections":3,"gates":"interval, count, unknown ids, confidence, then geometry (motion)"},"outputs":{"kept":[{"expression":"the man wearing a black cap who is talking to the camera","kind":"interaction","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":[[0.0,0.0,0.4791666865348816,0.9990740418434143],[0.0,0.0,0.47708335518836975,0.9990740418434143],[0.0,0.0,0.44583335518836975,0.9990740418434143],[0.0,0.0,0.4583333432674408,0.9990740418434143],[0.0,0.0,0.476041704416275,0.9990740418434143],[0.0,0.0,0.4833333492279053,0.9990740418434143],[0.0,0.0,0.5005208849906921,0.9990740418434143],[0.0,0.0,0.5104166865348816,0.9990740418434143],[0.0,0.0,0.5223958492279053,0.9990740418434143],[0.0,0.010185184888541698,0.47447919845581055,0.999074044637382],[0.0,0.0,0.5213541984558105,0.9990740418434143],[0.0,0.0,0.542187511920929,0.9990740418434143],[0.0,0.0,0.5562500357627869,0.9990740418434143],[0.0,0.0,0.5619791746139526,0.9990740418434143],[0.0,0.0,0.5687500238418579,0.9990740418434143],[0.0,0.0,0.5770833492279053,0.9990740418434143],[0.0,0.0,0.5692708492279053,0.9972221851348877],[0.0,0.0,0.5864583849906921,0.9990740418434143],[0.0,0.0,0.5739583373069763,0.9990740418434143],[0.0,0.0,0.5901042222976685,0.9990740418434143],[0.0,0.0,0.6276041865348816,0.9990740418434143],[0.0,0.0,0.6062500476837158,0.9990740418434143],[0.0,0.0,0.5911458730697632,0.9990740418434143],[0.0,0.0,0.5963541865348816,0.9990740418434143],[0.0,0.0,0.5661458373069763,0.9990740418434143],[0.0,0.0,0.5703125,0.9990740418434143],[0.0,0.0,0.5807291865348816,0.9990740418434143],[0.0,0.0,0.5963541865348816,0.9990740418434143],[0.0,0.0,0.9078125357627869,0.9990740418434143],[0.0,0.0,0.9218750596046448,0.9990740418434143],[0.0,0.0,0.9494792222976685,0.9990740418434143],[0.0,0.0,0.5869792103767395,0.9990740418434143],[0.0,0.0,0.5505208373069763,0.9990740418434143],[0.0,0.0,0.5729166865348816,0.9990740418434143],[0.0,0.0,0.5531250238418579,0.9990740418434143],[0.01927083358168602,0.0,0.5750000439584255,0.9990740418434143],[0.0,0.0,0.5786458849906921,0.9990740418434143],[0.0,0.0,0.5843750238418579,0.9990740418434143],[0.0,0.0,0.5723958611488342,0.9990740418434143],[0.0,0.0,0.5473958849906921,0.9990740418434143],[0.0,0.0,0.5666667222976685,0.9990740418434143],[0.0,0.0,0.5625,0.9990740418434143],[0.0,0.0,0.589062511920929,0.9990740418434143],[0.0,0.0,0.6145833730697632,0.9990740418434143],[0.0,0.0,0.6369792222976685,0.9990740418434143],[0.0,0.0,0.6427083611488342,0.9990740418434143],[0.0,0.0,0.6312500238418579,0.9990740418434143],[0.0,0.0,0.6333333849906921,0.9990740418434143],[0.0,0.0,0.6260417103767395,0.9972221851348877],[0.0,0.0,0.6229166984558105,0.9990740418434143],[0.0,0.0,0.6312500238418579,0.9990740418434143],[0.0,0.0,0.6348958611488342,0.9990740418434143],[0.0,0.0,0.6317708492279053,0.9990740418434143],[0.0,0.0,0.6302083730697632,0.9990740418434143],[0.0,0.0,0.6072916984558105,0.9990740418434143],[0.0,0.0,0.5864583849906921,0.9990740418434143],[0.0,0.0,0.5859375596046448,0.9990740418434143],[0.0,0.0,0.5921875238418579,0.9990740418434143],[0.0,0.0,0.5869792103767395,0.9990740418434143],[0.0,0.0,0.5828125476837158,0.9990740418434143],[0.0,0.0,0.5744792222976685,0.9990740418434143],[0.0,0.0,0.5708333849906921,0.9990740418434143],[0.0026041667442768812,0.0,0.5437500278931111,0.9990740418434143]]}]},{"expression":"the man sitting in the back 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":[[0.3281250298023224,0.24537037312984467,0.5427083671092987,0.9907407015562057],[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]]}]},{"expression":"the man in the back seat who is not wearing sunglasses","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/--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 wearing a black cap who is talking to the camera","kind":"interaction","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":[[0.0,0.0,0.4791666865348816,0.9990740418434143],[0.0,0.0,0.47708335518836975,0.9990740418434143],[0.0,0.0,0.44583335518836975,0.9990740418434143],[0.0,0.0,0.4583333432674408,0.9990740418434143],[0.0,0.0,0.476041704416275,0.9990740418434143],[0.0,0.0,0.4833333492279053,0.9990740418434143],[0.0,0.0,0.5005208849906921,0.9990740418434143],[0.0,0.0,0.5104166865348816,0.9990740418434143],[0.0,0.0,0.5223958492279053,0.9990740418434143],[0.0,0.010185184888541698,0.47447919845581055,0.999074044637382],[0.0,0.0,0.5213541984558105,0.9990740418434143],[0.0,0.0,0.542187511920929,0.9990740418434143],[0.0,0.0,0.5562500357627869,0.9990740418434143],[0.0,0.0,0.5619791746139526,0.9990740418434143],[0.0,0.0,0.5687500238418579,0.9990740418434143],[0.0,0.0,0.5770833492279053,0.9990740418434143],[0.0,0.0,0.5692708492279053,0.9972221851348877],[0.0,0.0,0.5864583849906921,0.9990740418434143],[0.0,0.0,0.5739583373069763,0.9990740418434143],[0.0,0.0,0.5901042222976685,0.9990740418434143],[0.0,0.0,0.6276041865348816,0.9990740418434143],[0.0,0.0,0.6062500476837158,0.9990740418434143],[0.0,0.0,0.5911458730697632,0.9990740418434143],[0.0,0.0,0.5963541865348816,0.9990740418434143],[0.0,0.0,0.5661458373069763,0.9990740418434143],[0.0,0.0,0.5703125,0.9990740418434143],[0.0,0.0,0.5807291865348816,0.9990740418434143],[0.0,0.0,0.5963541865348816,0.9990740418434143],[0.0,0.0,0.9078125357627869,0.9990740418434143],[0.0,0.0,0.9218750596046448,0.9990740418434143],[0.0,0.0,0.9494792222976685,0.9990740418434143],[0.0,0.0,0.5869792103767395,0.9990740418434143],[0.0,0.0,0.5505208373069763,0.9990740418434143],[0.0,0.0,0.5729166865348816,0.9990740418434143],[0.0,0.0,0.5531250238418579,0.9990740418434143],[0.01927083358168602,0.0,0.5750000439584255,0.9990740418434143],[0.0,0.0,0.5786458849906921,0.9990740418434143],[0.0,0.0,0.5843750238418579,0.9990740418434143],[0.0,0.0,0.5723958611488342,0.9990740418434143],[0.0,0.0,0.5473958849906921,0.9990740418434143],[0.0,0.0,0.5666667222976685,0.9990740418434143],[0.0,0.0,0.5625,0.9990740418434143],[0.0,0.0,0.589062511920929,0.9990740418434143],[0.0,0.0,0.6145833730697632,0.9990740418434143],[0.0,0.0,0.6369792222976685,0.9990740418434143],[0.0,0.0,0.6427083611488342,0.9990740418434143],[0.0,0.0,0.6312500238418579,0.9990740418434143],[0.0,0.0,0.6333333849906921,0.9990740418434143],[0.0,0.0,0.6260417103767395,0.9972221851348877],[0.0,0.0,0.6229166984558105,0.9990740418434143],[0.0,0.0,0.6312500238418579,0.9990740418434143],[0.0,0.0,0.6348958611488342,0.9990740418434143],[0.0,0.0,0.6317708492279053,0.9990740418434143],[0.0,0.0,0.6302083730697632,0.9990740418434143],[0.0,0.0,0.6072916984558105,0.9990740418434143],[0.0,0.0,0.5864583849906921,0.9990740418434143],[0.0,0.0,0.5859375596046448,0.9990740418434143],[0.0,0.0,0.5921875238418579,0.9990740418434143],[0.0,0.0,0.5869792103767395,0.9990740418434143],[0.0,0.0,0.5828125476837158,0.9990740418434143],[0.0,0.0,0.5744792222976685,0.9990740418434143],[0.0,0.0,0.5708333849906921,0.9990740418434143],[0.0026041667442768812,0.0,0.5437500278931111,0.9990740418434143]]}]},{"expression":"the man sitting in the back 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":[[0.3281250298023224,0.24537037312984467,0.5427083671092987,0.9907407015562057],[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]]}]},{"expression":"the man in the back seat who is not wearing sunglasses","kind":"appearance","is_negative":true,"tracks":[]}],"rmot_expression_count":3,"rmot_negative_count":1,"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","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 on the street\",\n      \"noun_phrase\": \"white pickup truck\",\n      \"kind\": \"spatial\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the green door on the right side of the street\",\n      \"noun_phrase\": \"green door\",\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 pickup truck parked on the street","noun_phrase":"white pickup truck","kind":"spatial","expected_count":1,"is_negative":false},{"expression":"the green door on the right side of the street","noun_phrase":"green door","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 pickup truck parked on the street","noun_phrase":"white pickup truck","kind":"spatial","expected_count":1,"is_negative":false},{"expression":"the green door on the right side of the street","noun_phrase":"green door","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":["green door","red car","white pickup truck"],"expressions_in":3,"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":3},"outputs":{"raw_tracks":[{"noun_phrase":"green door","object_id":0,"observations":[{"timestamp":1.04,"box":[0.6062500476837158,0.5916666388511658,0.6177083812654018,0.7722221910953522],"occluded":false},{"timestamp":1.3,"box":[0.6541666984558105,0.6379629373550415,0.6703125331550837,0.8324073851108551],"occluded":false},{"timestamp":1.56,"box":[0.6369792222976685,0.6259258985519409,0.6536458898335695,0.8314814567565918],"occluded":false},{"timestamp":1.82,"box":[0.5671875476837158,0.6138888597488403,0.5822917148470879,0.8101851493120193],"occluded":false},{"timestamp":2.08,"box":[0.5567708611488342,0.555555522441864,0.5729166958481073,0.7555555254220963],"occluded":false},{"timestamp":2.34,"box":[0.5885416865348816,0.5666666626930237,0.6062500216066837,0.7777777761220932],"occluded":false},{"timestamp":2.6,"box":[0.6088541746139526,0.5731481313705444,0.6286458428949118,0.7972221970558167],"occluded":false},{"timestamp":2.86,"box":[0.5718750357627869,0.6166666746139526,0.591666704043746,0.8462962955236435],"occluded":false},{"timestamp":3.12,"box":[0.5776041746139526,0.6129629611968994,0.5979166757315397,0.8546296209096909],"occluded":false},{"timestamp":3.38,"box":[0.5755208730697632,0.5833333134651184,0.5994792077690363,0.8324073851108551],"occluded":false},{"timestamp":3.64,"box":[0.5817708373069763,0.5796296000480652,0.6093750055879354,0.8379629254341125],"occluded":false},{"timestamp":3.9,"box":[0.5854166746139526,0.565740704536438,0.6156250089406967,0.8314814269542694],"occluded":false},{"timestamp":4.16,"box":[0.5526041984558105,0.5370370149612427,0.5843750350177288,0.8120370209217072],"occluded":false},{"timestamp":4.42,"box":[0.5953125357627869,0.5768518447875977,0.6270833723247051,0.8759259283542633],"occluded":false},{"timestamp":4.68,"box":[0.612500011920929,0.5638889074325562,0.6494791805744171,0.8805555701255798],"occluded":false},{"timestamp":4.94,"box":[0.6260417103767395,0.5944444537162781,0.6661458797752857,0.949999988079071],"occluded":false},{"timestamp":5.2,"box":[0.6421875357627869,0.5537036657333374,0.691145870834589,0.9324073493480682],"occluded":false},{"timestamp":5.46,"box":[0.6578125357627869,0.5546296238899231,0.7093750387430191,0.9666666388511658],"occluded":false},{"timestamp":5.72,"box":[0.7104166746139526,0.5916666388511658,0.7729166746139526,0.9722221791744232],"occluded":false},{"timestamp":5.98,"box":[0.7281250357627869,0.5731481313705444,0.804687537252903,0.9722222089767456],"occluded":false},{"timestamp":6.24,"box":[0.6843750476837158,0.5314814448356628,0.7619792148470879,0.9712962508201599],"occluded":false},{"timestamp":6.5,"box":[0.6895833611488342,0.5074074268341064,0.7880208641290665,0.9712963104248047],"occluded":false},{"timestamp":6.76,"box":[0.7562500238418579,0.5685185194015503,0.8869791924953461,0.9722222089767456],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.8114583492279053,0.5314814448356628,0.9776041954755783,0.9990740418434143],"occluded":false},{"timestamp":7.28,"box":[0.7776042222976685,0.5240740776062012,0.9682292342185974,0.9712962806224823],"occluded":false},{"timestamp":7.54,"box":[0.5604166984558105,0.46759259700775146,0.9994792342185974,0.9722222089767456],"occluded":false},{"timestamp":7.8,"box":[0.7270833849906921,0.4583333134651184,0.9994792342185974,0.9712962508201599],"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]},{"noun_phrase":"green door","object_id":1,"observations":[{"timestamp":2.08,"box":[0.5359375476837158,0.5490740537643433,0.5447917152196169,0.7148147970438004],"occluded":false},{"timestamp":2.34,"box":[0.565625011920929,0.5574073791503906,0.576562512665987,0.7351851463317871],"occluded":false},{"timestamp":2.6,"box":[0.5848958492279053,0.5694444179534912,0.5963541828095913,0.7490740418434143],"occluded":false},{"timestamp":2.86,"box":[0.546875,0.6083333492279053,0.5588541673496366,0.7925925999879837],"occluded":false},{"timestamp":3.12,"box":[0.5510417222976685,0.6092592477798462,0.5625000558793545,0.7972222119569778],"occluded":false},{"timestamp":3.38,"box":[0.5484375357627869,0.5768518447875977,0.5614583697170019,0.770370364189148],"occluded":false},{"timestamp":3.64,"box":[0.5536458492279053,0.5731481313705444,0.5692708492279053,0.7740740478038788],"occluded":false},{"timestamp":3.9,"box":[0.5552083849906921,0.555555522441864,0.5718750525265932,0.7638888508081436],"occluded":false},{"timestamp":4.16,"box":[0.5208333730697632,0.5435184836387634,0.5369792077690363,0.741666629910469],"occluded":false},{"timestamp":4.42,"box":[0.5588542222976685,0.5731481313705444,0.5750000569969416,0.7953703552484512],"occluded":false},{"timestamp":4.68,"box":[0.5744792222976685,0.5611110925674438,0.5921875573694706,0.79537034034729],"occluded":false},{"timestamp":4.94,"box":[0.5838541984558105,0.5907407402992249,0.6031250320374966,0.8481481373310089],"occluded":false},{"timestamp":5.2,"box":[0.5963541865348816,0.5490740537643433,0.6187500208616257,0.8185184895992279],"occluded":false},{"timestamp":5.46,"box":[0.6072916984558105,0.5518518686294556,0.6302083656191826,0.8361111283302307],"occluded":false},{"timestamp":5.72,"box":[0.6531250476837158,0.5888888835906982,0.6770833823829889,0.9046296179294586],"occluded":false},{"timestamp":5.98,"box":[0.6656250357627869,0.574999988079071,0.694791704416275,0.9083333015441895],"occluded":false},{"timestamp":6.24,"box":[0.6187500357627869,0.5379629731178284,0.648958370089531,0.8740740716457367],"occluded":false},{"timestamp":6.5,"box":[0.6192708611488342,0.4972222149372101,0.6572916954755783,0.8574073910713196],"occluded":false},{"timestamp":6.76,"box":[0.6703125238418579,0.5805555582046509,0.7109375260770321,0.9712962806224823],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.7046875357627869,0.5407407283782959,0.7520833723247051,0.970370352268219],"occluded":false},{"timestamp":7.28,"box":[0.6697916984558105,0.5601851940155029,0.7192708663642406,0.9712963104248047],"occluded":false},{"timestamp":7.54,"box":[0.6661458611488342,0.5222222208976746,0.7286458611488342,0.9712962806224823],"occluded":false},{"timestamp":7.8,"box":[0.7156250476837158,0.5287036895751953,0.9770833849906921,0.9712962806224823],"occluded":false},{"timestamp":8.06,"box":[0.7848958969116211,0.4564814865589142,0.99114590883255,0.9712962806224823],"occluded":false},{"timestamp":8.32,"box":[0.7645833492279053,0.43981480598449707,0.8911458551883698,0.9712962508201599],"occluded":false},{"timestamp":8.58,"box":[0.7406250238418579,0.192592591047287,0.999479204416275,0.9712962806224823],"occluded":false}],"scores":[0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781]},{"noun_phrase":"green door","object_id":2,"observations":[{"timestamp":2.34,"box":[0.5427083373069763,0.6037036776542664,0.5505208373069763,0.678703673183918],"occluded":false},{"timestamp":2.6,"box":[0.5593750476837158,0.6111111044883728,0.5687500480562449,0.6870370283722878],"occluded":false},{"timestamp":2.86,"box":[0.5203125476837158,0.6509259343147278,0.5296875480562449,0.727777786552906],"occluded":false},{"timestamp":3.12,"box":[0.5218750238418579,0.6518518328666687,0.5322916908189654,0.728703685104847],"occluded":false},{"timestamp":3.38,"box":[0.5182291865348816,0.6222221851348877,0.5291666872799397,0.699074037373066],"occluded":false},{"timestamp":3.64,"box":[0.5218750238418579,0.6203703880310059,0.534375024959445,0.6990740895271301],"occluded":false},{"timestamp":3.9,"box":[0.5218750238418579,0.604629635810852,0.534375024959445,0.6842592656612396],"occluded":false},{"timestamp":4.16,"box":[0.4828125238418579,0.5805555582046509,0.4963541915640235,0.6592592597007751],"occluded":false},{"timestamp":4.42,"box":[0.5161458849906921,0.6231481432914734,0.5307292193174362,0.706481471657753],"occluded":false},{"timestamp":4.68,"box":[0.5296875238418579,0.6157407164573669,0.5411458574235439,0.6999999731779099],"occluded":false},{"timestamp":4.94,"box":[0.5359375476837158,0.6527777910232544,0.5468750484287739,0.7407407537102699],"occluded":false},{"timestamp":5.2,"box":[0.5458333492279053,0.6203703880310059,0.5572916828095913,0.7055555731058121],"occluded":false},{"timestamp":5.46,"box":[0.5510417222976685,0.6240740418434143,0.5619792230427265,0.7120370045304298],"occluded":false},{"timestamp":5.72,"box":[0.5901042222976685,0.6731481552124023,0.600520889274776,0.7648148238658905],"occluded":false},{"timestamp":5.98,"box":[0.5989583730697632,0.6629629731178284,0.6093750400468707,0.7574074193835258],"occluded":false},{"timestamp":6.24,"box":[0.5489583611488342,0.6314814686775208,0.5609375284984708,0.7249999865889549],"occluded":false},{"timestamp":6.5,"box":[0.5453125238418579,0.5972222089767456,0.558333357796073,0.6925925761461258],"occluded":false},{"timestamp":6.76,"box":[0.5828125476837158,0.6990740895271301,0.5968750482425094,0.8009259402751923],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.6072916984558105,0.6601851582527161,0.6223958656191826,0.7648147866129875],"occluded":false},{"timestamp":7.28,"box":[0.5645833611488342,0.6935185194015503,0.5796875283122063,0.8009259253740311],"occluded":false},{"timestamp":7.54,"box":[0.5557292103767395,0.6666666269302368,0.5708333775401115,0.7722221836447716],"occluded":false},{"timestamp":7.8,"box":[0.5911458730697632,0.7064814567565918,0.6083333734422922,0.8185184895992279],"occluded":false},{"timestamp":8.06,"box":[0.6390625238418579,0.6759259104728699,0.658854192122817,0.7916666492819786],"occluded":false},{"timestamp":8.32,"box":[0.596875011920929,0.654629647731781,0.617187513038516,0.7638889029622078],"occluded":false},{"timestamp":8.58,"box":[0.6510416865348816,0.5166666507720947,0.6760416887700558,0.6324073895812035],"occluded":false},{"timestamp":8.84,"box":[0.7734375596046448,0.510185182094574,0.8078125603497028,0.6407407373189926],"occluded":false}],"scores":[0.7037037014961243,0.7037037014961243,0.7037037014961243,0.7037037014961243,0.7037037014961243,0.7037037014961243,0.7037037014961243,0.7037037014961243,0.7037037014961243,0.7037037014961243,0.7037037014961243,0.7037037014961243,0.7037037014961243,0.7037037014961243,0.7037037014961243,0.7037037014961243,0.7037037014961243,0.7037037014961243,0.7037037014961243,0.7037037014961243,0.7037037014961243,0.7037037014961243,0.7037037014961243,0.7037037014961243,0.7037037014961243,0.7037037014961243]},{"noun_phrase":"green door","object_id":3,"observations":[{"timestamp":4.68,"box":[0.35885417461395264,0.5231481194496155,0.37447917461395264,0.6324073746800423],"occluded":false},{"timestamp":4.94,"box":[0.3541666865348816,0.5592592358589172,0.37708335369825363,0.67314812541008],"occluded":false},{"timestamp":5.2,"box":[0.35572919249534607,0.5129629373550415,0.38645836152136326,0.6296296045184135],"occluded":false},{"timestamp":5.46,"box":[0.34843751788139343,0.5240740776062012,0.3864583522081375,0.6472222208976746],"occluded":false},{"timestamp":5.72,"box":[0.36927086114883423,0.5685185194015503,0.41979169845581055,0.6990740746259689],"occluded":false},{"timestamp":5.98,"box":[0.3734375238418579,0.5666666626930237,0.42239585891366005,0.6990740746259689],"occluded":false},{"timestamp":6.24,"box":[0.3192708492279053,0.5148147940635681,0.3734375201165676,0.652777761220932],"occluded":false},{"timestamp":6.5,"box":[0.31197917461395264,0.4722222089767456,0.3671875111758709,0.6138888746500015],"occluded":false},{"timestamp":6.76,"box":[0.33906251192092896,0.5685185194015503,0.39791668206453323,0.7129629552364349],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.3609375059604645,0.5305555462837219,0.4192708432674408,0.6796296238899231],"occluded":false},{"timestamp":7.28,"box":[0.30677086114883423,0.5629629492759705,0.3697917014360428,0.7194444239139557],"occluded":false},{"timestamp":7.54,"box":[0.2880208492279053,0.5462962985038757,0.35104168951511383,0.7101851850748062],"occluded":false},{"timestamp":7.8,"box":[0.31562501192092896,0.5824074149131775,0.38072918355464935,0.7490740716457367],"occluded":false},{"timestamp":8.06,"box":[0.35885417461395264,0.5490740537643433,0.42239584773778915,0.7194444239139557],"occluded":false},{"timestamp":8.32,"box":[0.3109375238418579,0.4972222149372101,0.3812500238418579,0.6805555522441864],"occluded":false},{"timestamp":8.58,"box":[0.3656250238418579,0.3592592477798462,0.43020836263895035,0.5833333134651184],"occluded":false},{"timestamp":8.84,"box":[0.32447919249534607,0.3305555582046509,0.530729204416275,0.5648148059844971],"occluded":false},{"timestamp":9.1,"box":[0.6343750357627869,0.2768518328666687,0.8104167133569717,0.9990740418434143],"occluded":false},{"timestamp":9.36,"box":[0.5375000238418579,0.2944444417953491,0.8104166984558105,0.5398148149251938],"occluded":false},{"timestamp":9.620000000000001,"box":[0.6078125238418579,0.23333333432674408,0.9130208790302277,0.5305555611848831],"occluded":false},{"timestamp":10.66,"box":[0.9067708849906921,0.25740739703178406,0.9994792267680168,0.5712962746620178],"occluded":false},{"timestamp":10.92,"box":[0.4802083671092987,0.24722221493721008,0.7578125596046448,0.5861110985279083],"occluded":false},{"timestamp":11.18,"box":[0.4947916865348816,0.31111109256744385,0.590625025331974,0.6351851522922516],"occluded":false},{"timestamp":11.44,"box":[0.3682291805744171,0.2481481432914734,0.46145835518836975,0.5898148119449615],"occluded":false},{"timestamp":11.700000000000001,"box":[0.2083333432674408,0.18611110746860504,0.33020834624767303,0.5851851850748062],"occluded":false},{"timestamp":11.96,"box":[0.12031250447034836,0.14907407760620117,0.2666666731238365,0.6120370328426361],"occluded":false},{"timestamp":12.22,"box":[0.0703125,0.2092592567205429,0.2317708432674408,0.7111111134290695],"occluded":false},{"timestamp":12.48,"box":[0.09947916865348816,0.3046296238899231,0.21302083879709244,0.7398147881031036],"occluded":false},{"timestamp":12.74,"box":[0.0,0.277777761220932,0.14270834624767303,0.8796295821666718],"occluded":false},{"timestamp":13.0,"box":[0.0,0.1518518477678299,0.1197916716337204,0.7879629284143448],"occluded":false},{"timestamp":13.26,"box":[0.0,0.10925925523042679,0.40781253576278687,0.9990740492939949],"occluded":false}],"scores":[0.719298243522644,0.719298243522644,0.719298243522644,0.719298243522644,0.719298243522644,0.719298243522644,0.719298243522644,0.719298243522644,0.719298243522644,0.719298243522644,0.719298243522644,0.719298243522644,0.719298243522644,0.719298243522644,0.719298243522644,0.719298243522644,0.719298243522644,0.719298243522644,0.719298243522644,0.719298243522644,0.719298243522644,0.719298243522644,0.719298243522644,0.719298243522644,0.719298243522644,0.719298243522644,0.719298243522644,0.719298243522644,0.719298243522644,0.719298243522644,0.719298243522644]},{"noun_phrase":"green door","object_id":4,"observations":[{"timestamp":11.18,"box":[0.9338542222976685,0.5768518447875977,0.9885417260229588,0.7564814686775208],"occluded":false},{"timestamp":11.44,"box":[0.7760416865348816,0.46666666865348816,0.8156250230967999,0.618518516421318],"occluded":false},{"timestamp":11.700000000000001,"box":[0.6401041746139526,0.4814814627170563,0.6739583425223827,0.6194444298744202],"occluded":false},{"timestamp":11.96,"box":[0.6041666865348816,0.4749999940395355,0.6364583559334278,0.6157407313585281],"occluded":false},{"timestamp":12.22,"box":[0.6041666865348816,0.510185182094574,0.6343750208616257,0.6564814746379852],"occluded":false},{"timestamp":12.48,"box":[0.6109375357627869,0.5212962627410889,0.641666704788804,0.6694444119930267],"occluded":false},{"timestamp":12.74,"box":[0.5713542103767395,0.6583333015441895,0.7270833849906921,0.9990740418434143],"occluded":false},{"timestamp":13.0,"box":[0.5854166746139526,0.5712962746620178,0.6192708425223827,0.7314814627170563],"occluded":false},{"timestamp":13.26,"box":[0.40781253576278687,0.5240740776062012,0.6375000476837158,0.9990740716457367],"occluded":false},{"timestamp":13.52,"box":[0.5828125476837158,0.5231481194496155,0.7968750596046448,0.9990740418434143],"occluded":false},{"timestamp":13.780000000000001,"box":[0.5807291865348816,0.5833333134651184,0.9078125357627869,0.9990740418434143],"occluded":false},{"timestamp":14.040000000000001,"box":[0.39375001192092896,0.54537034034729,0.6567708551883698,0.9990740418434143],"occluded":false},{"timestamp":14.3,"box":[0.6296875476837158,0.47777777910232544,0.682291716337204,0.9990740418434143],"occluded":false},{"timestamp":14.56,"box":[0.5979167222976685,0.4851851761341095,0.6390625573694706,0.6796296238899231],"occluded":false},{"timestamp":14.82,"box":[0.5697916746139526,0.5481481552124023,0.6156250089406967,0.7546296268701553],"occluded":false},{"timestamp":15.08,"box":[0.6109375357627869,0.4851851761341095,0.6588542051613331,0.7037036865949631],"occluded":false},{"timestamp":15.34,"box":[0.6546875238418579,0.46666666865348816,0.708333358168602,0.6990740746259689],"occluded":false},{"timestamp":15.6,"box":[0.6328125596046448,0.44814813137054443,0.688020896166563,0.6861110925674438],"occluded":false},{"timestamp":15.860000000000001,"box":[0.5791667103767395,0.46481481194496155,0.6359375454485416,0.7074074000120163],"occluded":false},{"timestamp":16.12,"box":[0.6062500476837158,0.4851851761341095,0.6687500476837158,0.7490740716457367],"occluded":false},{"timestamp":16.38,"box":[0.659375011920929,0.44814813137054443,0.7322916835546494,0.9990740418434143],"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]},{"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":"green door","object_id":2,"presentation_id":1,"observations":[{"timestamp":2.34,"box":[0.5427083373069763,0.6037036776542664,0.5505208373069763,0.678703673183918],"occluded":false},{"timestamp":2.6,"box":[0.5593750476837158,0.6111111044883728,0.5687500480562449,0.6870370283722878],"occluded":false},{"timestamp":2.86,"box":[0.5203125476837158,0.6509259343147278,0.5296875480562449,0.727777786552906],"occluded":false},{"timestamp":3.12,"box":[0.5218750238418579,0.6518518328666687,0.5322916908189654,0.728703685104847],"occluded":false},{"timestamp":3.38,"box":[0.5182291865348816,0.6222221851348877,0.5291666872799397,0.699074037373066],"occluded":false},{"timestamp":3.64,"box":[0.5218750238418579,0.6203703880310059,0.534375024959445,0.6990740895271301],"occluded":false},{"timestamp":3.9,"box":[0.5218750238418579,0.604629635810852,0.534375024959445,0.6842592656612396],"occluded":false},{"timestamp":4.16,"box":[0.4828125238418579,0.5805555582046509,0.4963541915640235,0.6592592597007751],"occluded":false},{"timestamp":4.42,"box":[0.5161458849906921,0.6231481432914734,0.5307292193174362,0.706481471657753],"occluded":false},{"timestamp":4.68,"box":[0.5296875238418579,0.6157407164573669,0.5411458574235439,0.6999999731779099],"occluded":false},{"timestamp":4.94,"box":[0.5359375476837158,0.6527777910232544,0.5468750484287739,0.7407407537102699],"occluded":false},{"timestamp":5.2,"box":[0.5458333492279053,0.6203703880310059,0.5572916828095913,0.7055555731058121],"occluded":false},{"timestamp":5.46,"box":[0.5510417222976685,0.6240740418434143,0.5619792230427265,0.7120370045304298],"occluded":false},{"timestamp":5.72,"box":[0.5901042222976685,0.6731481552124023,0.600520889274776,0.7648148238658905],"occluded":false},{"timestamp":5.98,"box":[0.5989583730697632,0.6629629731178284,0.6093750400468707,0.7574074193835258],"occluded":false},{"timestamp":6.24,"box":[0.5489583611488342,0.6314814686775208,0.5609375284984708,0.7249999865889549],"occluded":false},{"timestamp":6.5,"box":[0.5453125238418579,0.5972222089767456,0.558333357796073,0.6925925761461258],"occluded":false},{"timestamp":6.76,"box":[0.5828125476837158,0.6990740895271301,0.5968750482425094,0.8009259402751923],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.6072916984558105,0.6601851582527161,0.6223958656191826,0.7648147866129875],"occluded":false},{"timestamp":7.28,"box":[0.5645833611488342,0.6935185194015503,0.5796875283122063,0.8009259253740311],"occluded":false},{"timestamp":7.54,"box":[0.5557292103767395,0.6666666269302368,0.5708333775401115,0.7722221836447716],"occluded":false},{"timestamp":7.8,"box":[0.5911458730697632,0.7064814567565918,0.6083333734422922,0.8185184895992279],"occluded":false},{"timestamp":8.06,"box":[0.6390625238418579,0.6759259104728699,0.658854192122817,0.7916666492819786],"occluded":false},{"timestamp":8.32,"box":[0.596875011920929,0.654629647731781,0.617187513038516,0.7638889029622078],"occluded":false},{"timestamp":8.58,"box":[0.6510416865348816,0.5166666507720947,0.6760416887700558,0.6324073895812035],"occluded":false},{"timestamp":8.84,"box":[0.7734375596046448,0.510185182094574,0.8078125603497028,0.6407407373189926],"occluded":false}],"mean_score":0.7037037014961243},{"noun_phrase":"white pickup truck","object_id":0,"presentation_id":2,"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},{"noun_phrase":"green door","object_id":1,"presentation_id":3,"observations":[{"timestamp":2.08,"box":[0.5359375476837158,0.5490740537643433,0.5447917152196169,0.7148147970438004],"occluded":false},{"timestamp":2.34,"box":[0.565625011920929,0.5574073791503906,0.576562512665987,0.7351851463317871],"occluded":false},{"timestamp":2.6,"box":[0.5848958492279053,0.5694444179534912,0.5963541828095913,0.7490740418434143],"occluded":false},{"timestamp":2.86,"box":[0.546875,0.6083333492279053,0.5588541673496366,0.7925925999879837],"occluded":false},{"timestamp":3.12,"box":[0.5510417222976685,0.6092592477798462,0.5625000558793545,0.7972222119569778],"occluded":false},{"timestamp":3.38,"box":[0.5484375357627869,0.5768518447875977,0.5614583697170019,0.770370364189148],"occluded":false},{"timestamp":3.64,"box":[0.5536458492279053,0.5731481313705444,0.5692708492279053,0.7740740478038788],"occluded":false},{"timestamp":3.9,"box":[0.5552083849906921,0.555555522441864,0.5718750525265932,0.7638888508081436],"occluded":false},{"timestamp":4.16,"box":[0.5208333730697632,0.5435184836387634,0.5369792077690363,0.741666629910469],"occluded":false},{"timestamp":4.42,"box":[0.5588542222976685,0.5731481313705444,0.5750000569969416,0.7953703552484512],"occluded":false},{"timestamp":4.68,"box":[0.5744792222976685,0.5611110925674438,0.5921875573694706,0.79537034034729],"occluded":false},{"timestamp":4.94,"box":[0.5838541984558105,0.5907407402992249,0.6031250320374966,0.8481481373310089],"occluded":false},{"timestamp":5.2,"box":[0.5963541865348816,0.5490740537643433,0.6187500208616257,0.8185184895992279],"occluded":false},{"timestamp":5.46,"box":[0.6072916984558105,0.5518518686294556,0.6302083656191826,0.8361111283302307],"occluded":false},{"timestamp":5.72,"box":[0.6531250476837158,0.5888888835906982,0.6770833823829889,0.9046296179294586],"occluded":false},{"timestamp":5.98,"box":[0.6656250357627869,0.574999988079071,0.694791704416275,0.9083333015441895],"occluded":false},{"timestamp":6.24,"box":[0.6187500357627869,0.5379629731178284,0.648958370089531,0.8740740716457367],"occluded":false},{"timestamp":6.5,"box":[0.6192708611488342,0.4972222149372101,0.6572916954755783,0.8574073910713196],"occluded":false},{"timestamp":6.76,"box":[0.6703125238418579,0.5805555582046509,0.7109375260770321,0.9712962806224823],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.7046875357627869,0.5407407283782959,0.7520833723247051,0.970370352268219],"occluded":false},{"timestamp":7.28,"box":[0.6697916984558105,0.5601851940155029,0.7192708663642406,0.9712963104248047],"occluded":false},{"timestamp":7.54,"box":[0.6661458611488342,0.5222222208976746,0.7286458611488342,0.9712962806224823],"occluded":false},{"timestamp":7.8,"box":[0.7156250476837158,0.5287036895751953,0.9770833849906921,0.9712962806224823],"occluded":false},{"timestamp":8.06,"box":[0.7848958969116211,0.4564814865589142,0.99114590883255,0.9712962806224823],"occluded":false},{"timestamp":8.32,"box":[0.7645833492279053,0.43981480598449707,0.8911458551883698,0.9712962508201599],"occluded":false},{"timestamp":8.58,"box":[0.7406250238418579,0.192592591047287,0.999479204416275,0.9712962806224823],"occluded":false}],"mean_score":0.7104072570800781},{"noun_phrase":"green door","object_id":4,"presentation_id":4,"observations":[{"timestamp":11.18,"box":[0.9338542222976685,0.5768518447875977,0.9885417260229588,0.7564814686775208],"occluded":false},{"timestamp":11.44,"box":[0.7760416865348816,0.46666666865348816,0.8156250230967999,0.618518516421318],"occluded":false},{"timestamp":11.700000000000001,"box":[0.6401041746139526,0.4814814627170563,0.6739583425223827,0.6194444298744202],"occluded":false},{"timestamp":11.96,"box":[0.6041666865348816,0.4749999940395355,0.6364583559334278,0.6157407313585281],"occluded":false},{"timestamp":12.22,"box":[0.6041666865348816,0.510185182094574,0.6343750208616257,0.6564814746379852],"occluded":false},{"timestamp":12.48,"box":[0.6109375357627869,0.5212962627410889,0.641666704788804,0.6694444119930267],"occluded":false},{"timestamp":12.74,"box":[0.5713542103767395,0.6583333015441895,0.7270833849906921,0.9990740418434143],"occluded":false},{"timestamp":13.0,"box":[0.5854166746139526,0.5712962746620178,0.6192708425223827,0.7314814627170563],"occluded":false},{"timestamp":13.26,"box":[0.40781253576278687,0.5240740776062012,0.6375000476837158,0.9990740716457367],"occluded":false},{"timestamp":13.52,"box":[0.5828125476837158,0.5231481194496155,0.7968750596046448,0.9990740418434143],"occluded":false},{"timestamp":13.780000000000001,"box":[0.5807291865348816,0.5833333134651184,0.9078125357627869,0.9990740418434143],"occluded":false},{"timestamp":14.040000000000001,"box":[0.39375001192092896,0.54537034034729,0.6567708551883698,0.9990740418434143],"occluded":false},{"timestamp":14.3,"box":[0.6296875476837158,0.47777777910232544,0.682291716337204,0.9990740418434143],"occluded":false},{"timestamp":14.56,"box":[0.5979167222976685,0.4851851761341095,0.6390625573694706,0.6796296238899231],"occluded":false},{"timestamp":14.82,"box":[0.5697916746139526,0.5481481552124023,0.6156250089406967,0.7546296268701553],"occluded":false},{"timestamp":15.08,"box":[0.6109375357627869,0.4851851761341095,0.6588542051613331,0.7037036865949631],"occluded":false},{"timestamp":15.34,"box":[0.6546875238418579,0.46666666865348816,0.708333358168602,0.6990740746259689],"occluded":false},{"timestamp":15.6,"box":[0.6328125596046448,0.44814813137054443,0.688020896166563,0.6861110925674438],"occluded":false},{"timestamp":15.860000000000001,"box":[0.5791667103767395,0.46481481194496155,0.6359375454485416,0.7074074000120163],"occluded":false},{"timestamp":16.12,"box":[0.6062500476837158,0.4851851761341095,0.6687500476837158,0.7490740716457367],"occluded":false},{"timestamp":16.38,"box":[0.659375011920929,0.44814813137054443,0.7322916835546494,0.9990740418434143],"occluded":false}],"mean_score":0.9064327478408813},{"noun_phrase":"green door","object_id":0,"presentation_id":5,"observations":[{"timestamp":1.04,"box":[0.6062500476837158,0.5916666388511658,0.6177083812654018,0.7722221910953522],"occluded":false},{"timestamp":1.3,"box":[0.6541666984558105,0.6379629373550415,0.6703125331550837,0.8324073851108551],"occluded":false},{"timestamp":1.56,"box":[0.6369792222976685,0.6259258985519409,0.6536458898335695,0.8314814567565918],"occluded":false},{"timestamp":1.82,"box":[0.5671875476837158,0.6138888597488403,0.5822917148470879,0.8101851493120193],"occluded":false},{"timestamp":2.08,"box":[0.5567708611488342,0.555555522441864,0.5729166958481073,0.7555555254220963],"occluded":false},{"timestamp":2.34,"box":[0.5885416865348816,0.5666666626930237,0.6062500216066837,0.7777777761220932],"occluded":false},{"timestamp":2.6,"box":[0.6088541746139526,0.5731481313705444,0.6286458428949118,0.7972221970558167],"occluded":false},{"timestamp":2.86,"box":[0.5718750357627869,0.6166666746139526,0.591666704043746,0.8462962955236435],"occluded":false},{"timestamp":3.12,"box":[0.5776041746139526,0.6129629611968994,0.5979166757315397,0.8546296209096909],"occluded":false},{"timestamp":3.38,"box":[0.5755208730697632,0.5833333134651184,0.5994792077690363,0.8324073851108551],"occluded":false},{"timestamp":3.64,"box":[0.5817708373069763,0.5796296000480652,0.6093750055879354,0.8379629254341125],"occluded":false},{"timestamp":3.9,"box":[0.5854166746139526,0.565740704536438,0.6156250089406967,0.8314814269542694],"occluded":false},{"timestamp":4.16,"box":[0.5526041984558105,0.5370370149612427,0.5843750350177288,0.8120370209217072],"occluded":false},{"timestamp":4.42,"box":[0.5953125357627869,0.5768518447875977,0.6270833723247051,0.8759259283542633],"occluded":false},{"timestamp":4.68,"box":[0.612500011920929,0.5638889074325562,0.6494791805744171,0.8805555701255798],"occluded":false},{"timestamp":4.94,"box":[0.6260417103767395,0.5944444537162781,0.6661458797752857,0.949999988079071],"occluded":false},{"timestamp":5.2,"box":[0.6421875357627869,0.5537036657333374,0.691145870834589,0.9324073493480682],"occluded":false},{"timestamp":5.46,"box":[0.6578125357627869,0.5546296238899231,0.7093750387430191,0.9666666388511658],"occluded":false},{"timestamp":5.72,"box":[0.7104166746139526,0.5916666388511658,0.7729166746139526,0.9722221791744232],"occluded":false},{"timestamp":5.98,"box":[0.7281250357627869,0.5731481313705444,0.804687537252903,0.9722222089767456],"occluded":false},{"timestamp":6.24,"box":[0.6843750476837158,0.5314814448356628,0.7619792148470879,0.9712962508201599],"occluded":false},{"timestamp":6.5,"box":[0.6895833611488342,0.5074074268341064,0.7880208641290665,0.9712963104248047],"occluded":false},{"timestamp":6.76,"box":[0.7562500238418579,0.5685185194015503,0.8869791924953461,0.9722222089767456],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.8114583492279053,0.5314814448356628,0.9776041954755783,0.9990740418434143],"occluded":false},{"timestamp":7.28,"box":[0.7776042222976685,0.5240740776062012,0.9682292342185974,0.9712962806224823],"occluded":false},{"timestamp":7.54,"box":[0.5604166984558105,0.46759259700775146,0.9994792342185974,0.9722222089767456],"occluded":false},{"timestamp":7.8,"box":[0.7270833849906921,0.4583333134651184,0.9994792342185974,0.9712962508201599],"occluded":false}],"mean_score":0.7460317611694336},{"noun_phrase":"green door","object_id":3,"presentation_id":6,"observations":[{"timestamp":4.68,"box":[0.35885417461395264,0.5231481194496155,0.37447917461395264,0.6324073746800423],"occluded":false},{"timestamp":4.94,"box":[0.3541666865348816,0.5592592358589172,0.37708335369825363,0.67314812541008],"occluded":false},{"timestamp":5.2,"box":[0.35572919249534607,0.5129629373550415,0.38645836152136326,0.6296296045184135],"occluded":false},{"timestamp":5.46,"box":[0.34843751788139343,0.5240740776062012,0.3864583522081375,0.6472222208976746],"occluded":false},{"timestamp":5.72,"box":[0.36927086114883423,0.5685185194015503,0.41979169845581055,0.6990740746259689],"occluded":false},{"timestamp":5.98,"box":[0.3734375238418579,0.5666666626930237,0.42239585891366005,0.6990740746259689],"occluded":false},{"timestamp":6.24,"box":[0.3192708492279053,0.5148147940635681,0.3734375201165676,0.652777761220932],"occluded":false},{"timestamp":6.5,"box":[0.31197917461395264,0.4722222089767456,0.3671875111758709,0.6138888746500015],"occluded":false},{"timestamp":6.76,"box":[0.33906251192092896,0.5685185194015503,0.39791668206453323,0.7129629552364349],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.3609375059604645,0.5305555462837219,0.4192708432674408,0.6796296238899231],"occluded":false},{"timestamp":7.28,"box":[0.30677086114883423,0.5629629492759705,0.3697917014360428,0.7194444239139557],"occluded":false},{"timestamp":7.54,"box":[0.2880208492279053,0.5462962985038757,0.35104168951511383,0.7101851850748062],"occluded":false},{"timestamp":7.8,"box":[0.31562501192092896,0.5824074149131775,0.38072918355464935,0.7490740716457367],"occluded":false},{"timestamp":8.06,"box":[0.35885417461395264,0.5490740537643433,0.42239584773778915,0.7194444239139557],"occluded":false},{"timestamp":8.32,"box":[0.3109375238418579,0.4972222149372101,0.3812500238418579,0.6805555522441864],"occluded":false},{"timestamp":8.58,"box":[0.3656250238418579,0.3592592477798462,0.43020836263895035,0.5833333134651184],"occluded":false},{"timestamp":8.84,"box":[0.32447919249534607,0.3305555582046509,0.530729204416275,0.5648148059844971],"occluded":false},{"timestamp":9.1,"box":[0.6343750357627869,0.2768518328666687,0.8104167133569717,0.9990740418434143],"occluded":false},{"timestamp":9.36,"box":[0.5375000238418579,0.2944444417953491,0.8104166984558105,0.5398148149251938],"occluded":false},{"timestamp":9.620000000000001,"box":[0.6078125238418579,0.23333333432674408,0.9130208790302277,0.5305555611848831],"occluded":false},{"timestamp":10.66,"box":[0.9067708849906921,0.25740739703178406,0.9994792267680168,0.5712962746620178],"occluded":false},{"timestamp":10.92,"box":[0.4802083671092987,0.24722221493721008,0.7578125596046448,0.5861110985279083],"occluded":false},{"timestamp":11.18,"box":[0.4947916865348816,0.31111109256744385,0.590625025331974,0.6351851522922516],"occluded":false},{"timestamp":11.44,"box":[0.3682291805744171,0.2481481432914734,0.46145835518836975,0.5898148119449615],"occluded":false},{"timestamp":11.700000000000001,"box":[0.2083333432674408,0.18611110746860504,0.33020834624767303,0.5851851850748062],"occluded":false},{"timestamp":11.96,"box":[0.12031250447034836,0.14907407760620117,0.2666666731238365,0.6120370328426361],"occluded":false},{"timestamp":12.22,"box":[0.0703125,0.2092592567205429,0.2317708432674408,0.7111111134290695],"occluded":false},{"timestamp":12.48,"box":[0.09947916865348816,0.3046296238899231,0.21302083879709244,0.7398147881031036],"occluded":false},{"timestamp":12.74,"box":[0.0,0.277777761220932,0.14270834624767303,0.8796295821666718],"occluded":false},{"timestamp":13.0,"box":[0.0,0.1518518477678299,0.1197916716337204,0.7879629284143448],"occluded":false},{"timestamp":13.26,"box":[0.0,0.10925925523042679,0.40781253576278687,0.9990740492939949],"occluded":false}],"mean_score":0.719298243522644}],"dropped":{},"fates":[{"noun_phrase":"green door","object_id":0,"frames":27,"mean_score":0.7460317611694336,"mean_area":0.031434042951005624,"outcome":"kept","presentation_id":5,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"green door","object_id":1,"frames":26,"mean_score":0.7104072570800781,"mean_area":0.025535523948714667,"outcome":"kept","presentation_id":3,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"green door","object_id":2,"frames":26,"mean_score":0.7037037014961243,"mean_area":0.0013897050575906859,"outcome":"kept","presentation_id":1,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"green door","object_id":3,"frames":31,"mean_score":0.719298243522644,"mean_area":0.045755117200568074,"outcome":"kept","presentation_id":6,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"green door","object_id":4,"frames":21,"mean_score":0.9064327478408813,"mean_area":0.03381914700431279,"outcome":"kept","presentation_id":4,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"white pickup truck","object_id":0,"frames":13,"mean_score":0.7575758099555969,"mean_area":0.20784945087048315,"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":6,"as_video":true,"fps":1.0,"max_frames":30},"outputs":{"media":"/home/zzk/tmp/rmot_smoke_real_86260/renders/--cif7Nv6BI_chunk_011_scene_010_choices.mp4","media_bytes":5278979,"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","temperature":0.0,"as_video":true,"media_bytes":5278979,"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 on the street\n2. the green door on the right side of the street\n3. the red car parked on the street","questions":["the white pickup truck parked on the street","the green door on the right side of the street","the red car parked on the street"]},"outputs":{"raw_response":"{\n  \"answers\": [\n    {\n      \"description_index\": 1,\n      \"matching_ids\": [\n        2\n      ],\n      \"start_frame\": 10,\n      \"end_frame\": 13,\n      \"confidence\": 1.0\n    },\n    {\n      \"description_index\": 2,\n      \"matching_ids\": [\n        4,\n        5\n      ],\n      \"start_frame\": 2,\n      \"end_frame\": 17,\n      \"confidence\": 0.9\n    },\n    {\n      \"description_index\": 3,\n      \"matching_ids\": [],\n      \"start_frame\": 1,\n      \"end_frame\": 17,\n      \"confidence\": 1.0\n    }\n  ]\n}","raw_answers":[{"description_index":1,"matching_ids":[2],"start_frame":10,"end_frame":13,"confidence":1.0},{"description_index":2,"matching_ids":[4,5],"start_frame":2,"end_frame":17,"confidence":0.9},{"description_index":3,"matching_ids":[],"start_frame":1,"end_frame":17,"confidence":1.0}],"answers":[{"expression":"the white pickup truck parked on the street","matching_presentation_ids":[2],"start_choice_frame":10,"end_choice_frame":13,"confidence":1.0,"self_reported_confidence":1.0},{"expression":"the green door on the right side of the street","matching_presentation_ids":[4,5],"start_choice_frame":2,"end_choice_frame":17,"confidence":0.9,"self_reported_confidence":0.9},{"expression":"the red car parked on the street","matching_presentation_ids":[],"start_choice_frame":1,"end_choice_frame":17,"confidence":1.0,"self_reported_confidence":1.0}]}},"verify":{"inputs":{"expressions":3,"tracklets":6,"selections":3,"gates":"interval, count, unknown ids, confidence, then geometry (motion)"},"outputs":{"kept":[{"expression":"the white pickup truck parked on the street","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":{"the red car parked on the street":"no_candidates_retrieved","the green door on the right side of the street":"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_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 on the street","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","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 street lamp glowing in the distance down the alley\",\n      \"noun_phrase\": \"street lamp\",\n      \"kind\": \"spatial\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the decorative stone carving on the wall above the doorway\",\n      \"noun_phrase\": \"stone carving\",\n      \"kind\": \"appearance\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the person walking down the street\",\n      \"noun_phrase\": \"person\",\n      \"kind\": \"motion\",\n      \"expected_count\": 0,\n      \"is_negative\": true\n    }\n  ]\n}","raw_items":[{"expression":"the street lamp glowing in the distance down the alley","noun_phrase":"street lamp","kind":"spatial","expected_count":1,"is_negative":false},{"expression":"the decorative stone carving on the wall above the doorway","noun_phrase":"stone carving","kind":"appearance","expected_count":1,"is_negative":false},{"expression":"the person walking down the street","noun_phrase":"person","kind":"motion","expected_count":0,"is_negative":true}],"accepted":[{"expression":"the street lamp glowing in the distance down the alley","noun_phrase":"street lamp","kind":"spatial","expected_count":1,"is_negative":false},{"expression":"the decorative stone carving on the wall above the doorway","noun_phrase":"stone carving","kind":"appearance","expected_count":1,"is_negative":false},{"expression":"the person walking down the street","noun_phrase":"person","kind":"motion","expected_count":0,"is_negative":true}],"gated":[]}},"track":{"inputs":{"tracker":"Sam31MultiplexTracker","model_id":"/mnt/ceph3/ec/jacky/sam3.1","noun_phrases":["person","stone carving","street lamp"],"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 carving","object_id":0,"observations":[{"timestamp":0.78,"box":[0.2682291865348816,0.031481482088565826,0.3864583596587181,0.32685185223817825],"occluded":false},{"timestamp":1.04,"box":[0.3708333671092987,0.19629628956317902,0.484375037252903,0.5277777761220932],"occluded":false},{"timestamp":1.3,"box":[0.42135417461395264,0.28518518805503845,0.5437500178813934,0.6324073970317841],"occluded":false},{"timestamp":1.56,"box":[0.4270833432674408,0.25648146867752075,0.575520858168602,0.6601851582527161],"occluded":false},{"timestamp":1.82,"box":[0.42604169249534607,0.24351851642131805,0.5984375327825546,0.7037036865949631],"occluded":false},{"timestamp":2.08,"box":[0.4192708432674408,0.21574074029922485,0.6083333492279053,0.7037037014961243],"occluded":false},{"timestamp":2.34,"box":[0.29635417461395264,0.18518517911434174,0.6140625178813934,0.7129629701375961],"occluded":false},{"timestamp":2.6,"box":[0.2968750298023224,0.19907407462596893,0.6187500357627869,0.7194444388151169],"occluded":false},{"timestamp":2.86,"box":[0.29635417461395264,0.19907407462596893,0.616666704416275,0.724074050784111],"occluded":false},{"timestamp":3.12,"box":[0.31041666865348816,0.18518517911434174,0.612500011920929,0.7148148268461227],"occluded":false},{"timestamp":3.38,"box":[0.19947917759418488,0.11574073880910873,0.5875000208616257,0.6324073895812035],"occluded":false},{"timestamp":3.64,"box":[0.2828125059604645,0.046296294778585434,0.5489583611488342,0.5916666351258755],"occluded":false},{"timestamp":3.9,"box":[0.03281250223517418,0.031481482088565826,0.4947916828095913,0.4287037029862404],"occluded":false},{"timestamp":4.16,"box":[0.01875000074505806,0.028703702613711357,0.5885416753590107,0.8037036787718534],"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]},{"noun_phrase":"stone carving","object_id":1,"observations":[{"timestamp":0.78,"box":[0.3500000238418579,0.3462962806224823,0.47708337008953094,0.6296296119689941],"occluded":false},{"timestamp":1.04,"box":[0.4593750238418579,0.5185185074806213,0.6052083671092987,0.8046296238899231],"occluded":false},{"timestamp":1.3,"box":[0.5244792103767395,0.6129629611968994,0.6958333849906921,0.9203703701496124],"occluded":false},{"timestamp":1.56,"box":[0.5609375238418579,0.6259258985519409,0.772916704416275,0.9379629194736481],"occluded":false},{"timestamp":1.82,"box":[0.5885416865348816,0.6537036895751953,0.8437500298023224,0.9722222089767456],"occluded":false},{"timestamp":2.08,"box":[0.31041666865348816,0.17129629850387573,0.8812500536441803,0.9722222089767456],"occluded":false},{"timestamp":2.34,"box":[0.6083333492279053,0.6490740776062012,0.8916667103767395,0.9722222089767456],"occluded":false},{"timestamp":2.6,"box":[0.2968750298023224,0.25462961196899414,0.8921875655651093,0.9722222089767456],"occluded":false},{"timestamp":2.86,"box":[0.29635417461395264,0.19907407462596893,0.8609375357627869,0.9722221940755844],"occluded":false},{"timestamp":3.12,"box":[0.31041666865348816,0.24074073135852814,0.8364583551883698,0.9722222238779068],"occluded":false},{"timestamp":3.38,"box":[0.19947917759418488,0.40740740299224854,0.7973958998918533,0.9648147821426392],"occluded":false},{"timestamp":3.64,"box":[0.33802086114883423,0.3314814865589142,0.7145833671092987,0.9055555760860443],"occluded":false},{"timestamp":3.9,"box":[0.4505208432674408,0.3898147940635681,0.6177083551883698,0.710185170173645],"occluded":false},{"timestamp":4.16,"box":[0.37916669249534607,0.18703703582286835,0.5270833671092987,0.7898148149251938],"occluded":false},{"timestamp":4.42,"box":[0.15781250596046448,0.03055555559694767,0.5468750298023224,0.6231481526046991],"occluded":false},{"timestamp":4.68,"box":[0.12968750298023224,0.03055555559694767,0.3828125149011612,0.26203703321516514],"occluded":false},{"timestamp":4.94,"box":[0.1067708358168602,0.031481482088565826,0.2723958417773247,0.4287037029862404],"occluded":false}],"scores":[0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781,0.7104072570800781]},{"noun_phrase":"stone carving","object_id":2,"observations":[{"timestamp":1.56,"box":[0.17760416865348816,0.45277777314186096,0.4723958373069763,0.6370370239019394],"occluded":false},{"timestamp":1.82,"box":[0.14427083730697632,0.47592592239379883,0.4791666865348816,0.6907407343387604],"occluded":false},{"timestamp":2.08,"box":[0.1223958432674408,0.17129629850387573,0.47968754172325134,0.6972222328186035],"occluded":false},{"timestamp":2.34,"box":[0.12864583730697632,0.48055553436279297,0.48385417461395264,0.7101851552724838],"occluded":false},{"timestamp":2.6,"box":[0.13645833730697632,0.489814817905426,0.48906251788139343,0.7185185104608536],"occluded":false},{"timestamp":2.86,"box":[0.12968750298023224,0.19907407462596893,0.48333336412906647,0.7185184806585312],"occluded":false},{"timestamp":3.12,"box":[0.12604168057441711,0.47592592239379883,0.4776042103767395,0.6990740597248077],"occluded":false},{"timestamp":3.38,"box":[0.11927083879709244,0.11574073880910873,0.45625001937150955,0.6027777716517448],"occluded":false},{"timestamp":3.64,"box":[0.09427084028720856,0.08796296268701553,0.4218750149011612,0.5111111178994179],"occluded":false},{"timestamp":3.9,"box":[0.00364583358168602,0.032407406717538834,0.3671875260770321,0.33888888731598854],"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":"stone carving","object_id":3,"observations":[{"timestamp":1.82,"box":[0.2979166805744171,0.18425925076007843,0.3348958492279053,0.25555554777383804],"occluded":false},{"timestamp":2.08,"box":[0.28125,0.1657407432794571,0.3213541693985462,0.24351851642131805],"occluded":false},{"timestamp":2.34,"box":[0.2828125059604645,0.1796296238899231,0.3239583410322666,0.25740739703178406],"occluded":false},{"timestamp":2.6,"box":[0.28854167461395264,0.19074073433876038,0.32760417833924294,0.26851850748062134],"occluded":false},{"timestamp":2.86,"box":[0.28854167461395264,0.032407406717538834,0.3296875096857548,0.2638888843357563],"occluded":false},{"timestamp":3.12,"box":[0.29322919249534607,0.17037037014961243,0.33541669324040413,0.2481481432914734],"occluded":false},{"timestamp":3.38,"box":[0.29479166865348816,0.10555555671453476,0.3359375037252903,0.17407407611608505],"occluded":false},{"timestamp":3.64,"box":[0.27916666865348816,0.03333333134651184,0.3218750059604645,0.10092592239379883],"occluded":false}],"scores":[0.6995305418968201,0.6995305418968201,0.6995305418968201,0.6995305418968201,0.6995305418968201,0.6995305418968201,0.6995305418968201,0.6995305418968201]},{"noun_phrase":"stone carving","object_id":4,"observations":[{"timestamp":2.34,"box":[0.8197917342185974,0.4416666626930237,0.992187574505806,0.64907406270504],"occluded":false},{"timestamp":2.6,"box":[0.8260416984558105,0.4462962746620178,0.999479204416275,0.6379629373550415],"occluded":false},{"timestamp":2.86,"box":[0.8244792222976685,0.46203702688217163,0.9963542371988297,0.662962943315506],"occluded":false},{"timestamp":3.12,"box":[0.8213542103767395,0.47685185074806213,0.9875000566244125,0.692592591047287],"occluded":false},{"timestamp":3.38,"box":[0.7817708849906921,0.44907405972480774,0.9182292222976685,0.6472222059965134],"occluded":false},{"timestamp":3.64,"box":[0.7234375476837158,0.4185185134410858,0.8348958864808083,0.6055555492639542],"occluded":false},{"timestamp":3.9,"box":[0.6505208611488342,0.34166666865348816,0.7343750298023224,0.5009259283542633],"occluded":false},{"timestamp":4.16,"box":[0.43541669845581055,0.2231481373310089,0.6416667103767395,0.7898148000240326],"occluded":false},{"timestamp":4.42,"box":[0.5317708849906921,0.08796296268701553,0.5854167193174362,0.24259259551763535],"occluded":false},{"timestamp":4.68,"box":[0.4755208492279053,0.031481482088565826,0.5291666835546494,0.14351851493120193],"occluded":false},{"timestamp":5.72,"box":[0.41197919845581055,0.11296296119689941,0.43229169957339764,0.1935185119509697],"occluded":false},{"timestamp":5.98,"box":[0.3890625238418579,0.11574073880910873,0.409375024959445,0.2074074074625969],"occluded":false},{"timestamp":6.24,"box":[0.37968751788139343,0.03981481492519379,0.4494791850447655,0.22685185074806213],"occluded":false},{"timestamp":6.5,"box":[0.37552085518836975,0.03888888657093048,0.44947919249534607,0.2074074000120163],"occluded":false},{"timestamp":6.76,"box":[0.3609375059604645,0.06666666269302368,0.43541667610406876,0.25648146867752075],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.3531250059604645,0.14722222089767456,0.37968750670552254,0.25],"occluded":false},{"timestamp":7.28,"box":[0.34427085518836975,0.09351851791143417,0.3718750234693289,0.19999999552965164],"occluded":false},{"timestamp":7.54,"box":[0.33385419845581055,0.08703703433275223,0.3640625327825546,0.21574073284864426],"occluded":false},{"timestamp":7.8,"box":[0.3255208432674408,0.05370370298624039,0.35729167982935905,0.1962962970137596],"occluded":false},{"timestamp":8.06,"box":[0.3489583432674408,0.032407406717538834,0.38333334401249886,0.16388889029622078],"occluded":false},{"timestamp":8.32,"box":[0.33645835518836975,0.06111111119389534,0.37187502533197403,0.18981480970978737],"occluded":false},{"timestamp":8.58,"box":[0.3281250298023224,0.031481482088565826,0.36614586412906647,0.1527777761220932],"occluded":false},{"timestamp":8.84,"box":[0.33020836114883423,0.031481482088565826,0.35572919622063637,0.09074074029922485],"occluded":false},{"timestamp":9.1,"box":[0.33020836114883423,0.031481482088565826,0.3510416951030493,0.056481482461094856],"occluded":false}],"scores":[0.719298243522644,0.719298243522644,0.719298243522644,0.719298243522644,0.719298243522644,0.719298243522644,0.719298243522644,0.719298243522644,0.719298243522644,0.719298243522644,0.719298243522644,0.719298243522644,0.719298243522644,0.719298243522644,0.719298243522644,0.719298243522644,0.719298243522644,0.719298243522644,0.719298243522644,0.719298243522644,0.719298243522644,0.719298243522644,0.719298243522644,0.719298243522644]},{"noun_phrase":"stone carving","object_id":5,"observations":[{"timestamp":5.46,"box":[0.00729166716337204,0.1787036955356598,0.43541669100522995,0.5879629552364349],"occluded":false},{"timestamp":5.72,"box":[0.16979166865348816,0.17777776718139648,0.3593750149011612,0.44999998807907104],"occluded":false},{"timestamp":5.98,"box":[0.10468750447034836,0.16388888657093048,0.330729179084301,0.4620370417833328],"occluded":false}],"scores":[0.6559139490127563,0.6559139490127563,null]},{"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]}],"tracklets":[{"noun_phrase":"street lamp","object_id":1,"presentation_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}],"mean_score":0.9064327478408813},{"noun_phrase":"street lamp","object_id":0,"presentation_id":2,"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":"stone carving","object_id":4,"presentation_id":3,"observations":[{"timestamp":2.34,"box":[0.8197917342185974,0.4416666626930237,0.992187574505806,0.64907406270504],"occluded":false},{"timestamp":2.6,"box":[0.8260416984558105,0.4462962746620178,0.999479204416275,0.6379629373550415],"occluded":false},{"timestamp":2.86,"box":[0.8244792222976685,0.46203702688217163,0.9963542371988297,0.662962943315506],"occluded":false},{"timestamp":3.12,"box":[0.8213542103767395,0.47685185074806213,0.9875000566244125,0.692592591047287],"occluded":false},{"timestamp":3.38,"box":[0.7817708849906921,0.44907405972480774,0.9182292222976685,0.6472222059965134],"occluded":false},{"timestamp":3.64,"box":[0.7234375476837158,0.4185185134410858,0.8348958864808083,0.6055555492639542],"occluded":false},{"timestamp":3.9,"box":[0.6505208611488342,0.34166666865348816,0.7343750298023224,0.5009259283542633],"occluded":false},{"timestamp":4.16,"box":[0.43541669845581055,0.2231481373310089,0.6416667103767395,0.7898148000240326],"occluded":false},{"timestamp":4.42,"box":[0.5317708849906921,0.08796296268701553,0.5854167193174362,0.24259259551763535],"occluded":false},{"timestamp":4.68,"box":[0.4755208492279053,0.031481482088565826,0.5291666835546494,0.14351851493120193],"occluded":false},{"timestamp":5.72,"box":[0.41197919845581055,0.11296296119689941,0.43229169957339764,0.1935185119509697],"occluded":false},{"timestamp":5.98,"box":[0.3890625238418579,0.11574073880910873,0.409375024959445,0.2074074074625969],"occluded":false},{"timestamp":6.24,"box":[0.37968751788139343,0.03981481492519379,0.4494791850447655,0.22685185074806213],"occluded":false},{"timestamp":6.5,"box":[0.37552085518836975,0.03888888657093048,0.44947919249534607,0.2074074000120163],"occluded":false},{"timestamp":6.76,"box":[0.3609375059604645,0.06666666269302368,0.43541667610406876,0.25648146867752075],"occluded":false},{"timestamp":7.0200000000000005,"box":[0.3531250059604645,0.14722222089767456,0.37968750670552254,0.25],"occluded":false},{"timestamp":7.28,"box":[0.34427085518836975,0.09351851791143417,0.3718750234693289,0.19999999552965164],"occluded":false},{"timestamp":7.54,"box":[0.33385419845581055,0.08703703433275223,0.3640625327825546,0.21574073284864426],"occluded":false},{"timestamp":7.8,"box":[0.3255208432674408,0.05370370298624039,0.35729167982935905,0.1962962970137596],"occluded":false},{"timestamp":8.06,"box":[0.3489583432674408,0.032407406717538834,0.38333334401249886,0.16388889029622078],"occluded":false},{"timestamp":8.32,"box":[0.33645835518836975,0.06111111119389534,0.37187502533197403,0.18981480970978737],"occluded":false},{"timestamp":8.58,"box":[0.3281250298023224,0.031481482088565826,0.36614586412906647,0.1527777761220932],"occluded":false},{"timestamp":8.84,"box":[0.33020836114883423,0.031481482088565826,0.35572919622063637,0.09074074029922485],"occluded":false},{"timestamp":9.1,"box":[0.33020836114883423,0.031481482088565826,0.3510416951030493,0.056481482461094856],"occluded":false}],"mean_score":0.719298243522644},{"noun_phrase":"stone carving","object_id":3,"presentation_id":4,"observations":[{"timestamp":1.82,"box":[0.2979166805744171,0.18425925076007843,0.3348958492279053,0.25555554777383804],"occluded":false},{"timestamp":2.08,"box":[0.28125,0.1657407432794571,0.3213541693985462,0.24351851642131805],"occluded":false},{"timestamp":2.34,"box":[0.2828125059604645,0.1796296238899231,0.3239583410322666,0.25740739703178406],"occluded":false},{"timestamp":2.6,"box":[0.28854167461395264,0.19074073433876038,0.32760417833924294,0.26851850748062134],"occluded":false},{"timestamp":2.86,"box":[0.28854167461395264,0.032407406717538834,0.3296875096857548,0.2638888843357563],"occluded":false},{"timestamp":3.12,"box":[0.29322919249534607,0.17037037014961243,0.33541669324040413,0.2481481432914734],"occluded":false},{"timestamp":3.38,"box":[0.29479166865348816,0.10555555671453476,0.3359375037252903,0.17407407611608505],"occluded":false},{"timestamp":3.64,"box":[0.27916666865348816,0.03333333134651184,0.3218750059604645,0.10092592239379883],"occluded":false}],"mean_score":0.6995305418968201},{"noun_phrase":"stone carving","object_id":0,"presentation_id":5,"observations":[{"timestamp":0.78,"box":[0.2682291865348816,0.031481482088565826,0.3864583596587181,0.32685185223817825],"occluded":false},{"timestamp":1.04,"box":[0.3708333671092987,0.19629628956317902,0.484375037252903,0.5277777761220932],"occluded":false},{"timestamp":1.3,"box":[0.42135417461395264,0.28518518805503845,0.5437500178813934,0.6324073970317841],"occluded":false},{"timestamp":1.56,"box":[0.4270833432674408,0.25648146867752075,0.575520858168602,0.6601851582527161],"occluded":false},{"timestamp":1.82,"box":[0.42604169249534607,0.24351851642131805,0.5984375327825546,0.7037036865949631],"occluded":false},{"timestamp":2.08,"box":[0.4192708432674408,0.21574074029922485,0.6083333492279053,0.7037037014961243],"occluded":false},{"timestamp":2.34,"box":[0.29635417461395264,0.18518517911434174,0.6140625178813934,0.7129629701375961],"occluded":false},{"timestamp":2.6,"box":[0.2968750298023224,0.19907407462596893,0.6187500357627869,0.7194444388151169],"occluded":false},{"timestamp":2.86,"box":[0.29635417461395264,0.19907407462596893,0.616666704416275,0.724074050784111],"occluded":false},{"timestamp":3.12,"box":[0.31041666865348816,0.18518517911434174,0.612500011920929,0.7148148268461227],"occluded":false},{"timestamp":3.38,"box":[0.19947917759418488,0.11574073880910873,0.5875000208616257,0.6324073895812035],"occluded":false},{"timestamp":3.64,"box":[0.2828125059604645,0.046296294778585434,0.5489583611488342,0.5916666351258755],"occluded":false},{"timestamp":3.9,"box":[0.03281250223517418,0.031481482088565826,0.4947916828095913,0.4287037029862404],"occluded":false},{"timestamp":4.16,"box":[0.01875000074505806,0.028703702613711357,0.5885416753590107,0.8037036787718534],"occluded":false}],"mean_score":0.7852349281311035},{"noun_phrase":"stone carving","object_id":5,"presentation_id":6,"observations":[{"timestamp":5.46,"box":[0.00729166716337204,0.1787036955356598,0.43541669100522995,0.5879629552364349],"occluded":false},{"timestamp":5.72,"box":[0.16979166865348816,0.17777776718139648,0.3593750149011612,0.44999998807907104],"occluded":false},{"timestamp":5.98,"box":[0.10468750447034836,0.16388888657093048,0.330729179084301,0.4620370417833328],"occluded":false}],"mean_score":0.6559139490127563},{"noun_phrase":"stone carving","object_id":2,"presentation_id":7,"observations":[{"timestamp":1.56,"box":[0.17760416865348816,0.45277777314186096,0.4723958373069763,0.6370370239019394],"occluded":false},{"timestamp":1.82,"box":[0.14427083730697632,0.47592592239379883,0.4791666865348816,0.6907407343387604],"occluded":false},{"timestamp":2.08,"box":[0.1223958432674408,0.17129629850387573,0.47968754172325134,0.6972222328186035],"occluded":false},{"timestamp":2.34,"box":[0.12864583730697632,0.48055553436279297,0.48385417461395264,0.7101851552724838],"occluded":false},{"timestamp":2.6,"box":[0.13645833730697632,0.489814817905426,0.48906251788139343,0.7185185104608536],"occluded":false},{"timestamp":2.86,"box":[0.12968750298023224,0.19907407462596893,0.48333336412906647,0.7185184806585312],"occluded":false},{"timestamp":3.12,"box":[0.12604168057441711,0.47592592239379883,0.4776042103767395,0.6990740597248077],"occluded":false},{"timestamp":3.38,"box":[0.11927083879709244,0.11574073880910873,0.45625001937150955,0.6027777716517448],"occluded":false},{"timestamp":3.64,"box":[0.09427084028720856,0.08796296268701553,0.4218750149011612,0.5111111178994179],"occluded":false},{"timestamp":3.9,"box":[0.00364583358168602,0.032407406717538834,0.3671875260770321,0.33888888731598854],"occluded":false}],"mean_score":0.6521739363670349},{"noun_phrase":"stone carving","object_id":1,"presentation_id":8,"observations":[{"timestamp":0.78,"box":[0.3500000238418579,0.3462962806224823,0.47708337008953094,0.6296296119689941],"occluded":false},{"timestamp":1.04,"box":[0.4593750238418579,0.5185185074806213,0.6052083671092987,0.8046296238899231],"occluded":false},{"timestamp":1.3,"box":[0.5244792103767395,0.6129629611968994,0.6958333849906921,0.9203703701496124],"occluded":false},{"timestamp":1.56,"box":[0.5609375238418579,0.6259258985519409,0.772916704416275,0.9379629194736481],"occluded":false},{"timestamp":1.82,"box":[0.5885416865348816,0.6537036895751953,0.8437500298023224,0.9722222089767456],"occluded":false},{"timestamp":2.08,"box":[0.31041666865348816,0.17129629850387573,0.8812500536441803,0.9722222089767456],"occluded":false},{"timestamp":2.34,"box":[0.6083333492279053,0.6490740776062012,0.8916667103767395,0.9722222089767456],"occluded":false},{"timestamp":2.6,"box":[0.2968750298023224,0.25462961196899414,0.8921875655651093,0.9722222089767456],"occluded":false},{"timestamp":2.86,"box":[0.29635417461395264,0.19907407462596893,0.8609375357627869,0.9722221940755844],"occluded":false},{"timestamp":3.12,"box":[0.31041666865348816,0.24074073135852814,0.8364583551883698,0.9722222238779068],"occluded":false},{"timestamp":3.38,"box":[0.19947917759418488,0.40740740299224854,0.7973958998918533,0.9648147821426392],"occluded":false},{"timestamp":3.64,"box":[0.33802086114883423,0.3314814865589142,0.7145833671092987,0.9055555760860443],"occluded":false},{"timestamp":3.9,"box":[0.4505208432674408,0.3898147940635681,0.6177083551883698,0.710185170173645],"occluded":false},{"timestamp":4.16,"box":[0.37916669249534607,0.18703703582286835,0.5270833671092987,0.7898148149251938],"occluded":false},{"timestamp":4.42,"box":[0.15781250596046448,0.03055555559694767,0.5468750298023224,0.6231481526046991],"occluded":false},{"timestamp":4.68,"box":[0.12968750298023224,0.03055555559694767,0.3828125149011612,0.26203703321516514],"occluded":false},{"timestamp":4.94,"box":[0.1067708358168602,0.031481482088565826,0.2723958417773247,0.4287037029862404],"occluded":false}],"mean_score":0.7104072570800781}],"dropped":{},"fates":[{"noun_phrase":"stone carving","object_id":0,"frames":14,"mean_score":0.7852349281311035,"mean_area":0.14147311396194637,"outcome":"kept","presentation_id":5,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"stone carving","object_id":1,"frames":17,"mean_score":0.7104072570800781,"mean_area":0.18365900527990892,"outcome":"kept","presentation_id":8,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"stone carving","object_id":2,"frames":10,"mean_score":0.6521739363670349,"mean_area":0.11526910116611448,"outcome":"kept","presentation_id":7,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"stone carving","object_id":3,"frames":8,"mean_score":0.6995305418968201,"mean_area":0.0038132355249246783,"outcome":"kept","presentation_id":4,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"stone carving","object_id":4,"frames":24,"mean_score":0.719298243522644,"mean_area":0.016865034031059142,"outcome":"kept","presentation_id":3,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"stone carving","object_id":5,"frames":3,"mean_score":0.6559139490127563,"mean_area":0.09807227938830732,"outcome":"kept","presentation_id":6,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"street lamp","object_id":0,"frames":45,"mean_score":0.6908212304115295,"mean_area":0.0036833420560621073,"outcome":"kept","presentation_id":2,"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":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":8,"as_video":true,"fps":1.0,"max_frames":30},"outputs":{"media":"/home/zzk/tmp/rmot_smoke_real_86260/renders/--cif7Nv6BI_chunk_012_scene_020_choices.mp4","media_bytes":3577522,"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","temperature":0.0,"as_video":true,"media_bytes":3577522,"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 street lamp glowing in the distance down the alley\n2. the decorative stone carving on the wall above the doorway\n3. the person walking down the street","questions":["the street lamp glowing in the distance down the alley","the decorative stone carving on the wall above the doorway","the person walking down the street"]},"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        5\n      ],\n      \"start_frame\": 2,\n      \"end_frame\": 4,\n      \"confidence\": 0.95\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":[2],"start_frame":1,"end_frame":16,"confidence":1.0},{"description_index":2,"matching_ids":[5],"start_frame":2,"end_frame":4,"confidence":0.95},{"description_index":3,"matching_ids":[],"start_frame":1,"end_frame":16,"confidence":1.0}],"answers":[{"expression":"the street lamp glowing in the distance down the alley","matching_presentation_ids":[2],"start_choice_frame":1,"end_choice_frame":16,"confidence":1.0,"self_reported_confidence":1.0},{"expression":"the decorative stone carving on the wall above the doorway","matching_presentation_ids":[5],"start_choice_frame":2,"end_choice_frame":4,"confidence":0.95,"self_reported_confidence":0.95},{"expression":"the person walking down 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":8,"selections":3,"gates":"interval, count, unknown ids, confidence, then geometry (motion)"},"outputs":{"kept":[{"expression":"the decorative stone carving on the wall above the doorway","kind":"appearance","is_negative":false,"tracks":[{"timestamps":[1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86],"boxes":[[0.3708333671092987,0.19629628956317902,0.484375037252903,0.5277777761220932],[0.42135417461395264,0.28518518805503845,0.5437500178813934,0.6324073970317841],[0.4270833432674408,0.25648146867752075,0.575520858168602,0.6601851582527161],[0.42604169249534607,0.24351851642131805,0.5984375327825546,0.7037036865949631],[0.4192708432674408,0.21574074029922485,0.6083333492279053,0.7037037014961243],[0.29635417461395264,0.18518517911434174,0.6140625178813934,0.7129629701375961],[0.2968750298023224,0.19907407462596893,0.6187500357627869,0.7194444388151169],[0.29635417461395264,0.19907407462596893,0.616666704416275,0.724074050784111]]}]}],"rejected":{"the person walking down the street":"no_candidates_retrieved","the street lamp glowing in the distance down the alley":"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 decorative stone carving on the wall above the doorway","kind":"appearance","is_negative":false,"tracks":[{"timestamps":[1.04,1.3,1.56,1.82,2.08,2.34,2.6,2.86],"boxes":[[0.3708333671092987,0.19629628956317902,0.484375037252903,0.5277777761220932],[0.42135417461395264,0.28518518805503845,0.5437500178813934,0.6324073970317841],[0.4270833432674408,0.25648146867752075,0.575520858168602,0.6601851582527161],[0.42604169249534607,0.24351851642131805,0.5984375327825546,0.7037036865949631],[0.4192708432674408,0.21574074029922485,0.6083333492279053,0.7037037014961243],[0.29635417461395264,0.18518517911434174,0.6140625178813934,0.7129629701375961],[0.2968750298023224,0.19907407462596893,0.6187500357627869,0.7194444388151169],[0.29635417461395264,0.19907407462596893,0.616666704416275,0.724074050784111]]}]}],"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","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 woman with long dark hair who briefly appears on the left\",\n      \"noun_phrase\": \"woman with long dark hair\",\n      \"kind\": \"appearance\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the glowing streetlamp mounted on the left wall\",\n      \"noun_phrase\": \"glowing streetlamp\",\n      \"kind\": \"spatial\",\n      \"expected_count\": 1,\n      \"is_negative\": false\n    },\n    {\n      \"expression\": \"the black car driving down the cobblestone street\",\n      \"noun_phrase\": \"black car\",\n      \"kind\": \"motion\",\n      \"expected_count\": 0,\n      \"is_negative\": true\n    }\n  ]\n}","raw_items":[{"expression":"the woman with long dark hair who briefly appears on the left","noun_phrase":"woman with long dark hair","kind":"appearance","expected_count":1,"is_negative":false},{"expression":"the glowing streetlamp mounted on the left wall","noun_phrase":"glowing streetlamp","kind":"spatial","expected_count":1,"is_negative":false},{"expression":"the black car driving down the cobblestone street","noun_phrase":"black car","kind":"motion","expected_count":0,"is_negative":true}],"accepted":[{"expression":"the woman with long dark hair who briefly appears on the left","noun_phrase":"woman with long dark hair","kind":"appearance","expected_count":1,"is_negative":false},{"expression":"the glowing streetlamp mounted on the left wall","noun_phrase":"glowing streetlamp","kind":"spatial","expected_count":1,"is_negative":false},{"expression":"the black car driving down the cobblestone street","noun_phrase":"black car","kind":"motion","expected_count":0,"is_negative":true}],"gated":[]}},"track":{"inputs":{"tracker":"Sam31MultiplexTracker","model_id":"/mnt/ceph3/ec/jacky/sam3.1","noun_phrases":["black car","glowing streetlamp","woman with long dark hair"],"expressions_in":3,"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":3},"outputs":{"raw_tracks":[{"noun_phrase":"glowing streetlamp","object_id":0,"observations":[{"timestamp":0.78,"box":[0.06458333879709244,0.031481482088565826,0.13854167610406876,0.14537037163972855],"occluded":false},{"timestamp":1.04,"box":[0.15781250596046448,0.03055555559694767,0.22604167461395264,0.12592592276632786],"occluded":false},{"timestamp":1.3,"box":[0.1067708358168602,0.03055555559694767,0.18177083879709244,0.09166666679084301],"occluded":false},{"timestamp":1.56,"box":[0.0520833358168602,0.03055555559694767,0.11093750596046448,0.06574073992669582],"occluded":false},{"timestamp":1.82,"box":[0.02968750149011612,0.03055555559694767,0.12083333730697632,0.13611111231148243],"occluded":false},{"timestamp":2.08,"box":[0.05104167014360428,0.03055555559694767,0.14114584028720856,0.14351851679384708],"occluded":false},{"timestamp":2.34,"box":[0.0911458358168602,0.03055555559694767,0.17656251043081284,0.3240740690380335],"occluded":false},{"timestamp":2.6,"box":[0.12708334624767303,0.04722222313284874,0.20781251788139343,0.17777777835726738],"occluded":false},{"timestamp":2.86,"box":[0.10572917014360428,0.09074074029922485,0.19739583879709244,0.43518516421318054],"occluded":false},{"timestamp":3.12,"box":[0.07447917014360428,0.08425925672054291,0.17083334177732468,0.4287036806344986],"occluded":false},{"timestamp":3.38,"box":[0.08437500149011612,0.06203703582286835,0.17500000447034836,0.21111111342906952],"occluded":false},{"timestamp":3.64,"box":[0.10208334028720856,0.05092592537403107,0.20000001043081284,0.3453703671693802],"occluded":false},{"timestamp":3.9,"box":[0.09791667014360428,0.032407406717538834,0.18697917461395264,0.18703703954815865],"occluded":false},{"timestamp":4.16,"box":[0.07395833730697632,0.03333333134651184,0.16822917759418488,0.18518517911434174],"occluded":false},{"timestamp":4.42,"box":[0.04010416939854622,0.031481482088565826,0.1432291753590107,0.1787037029862404],"occluded":false},{"timestamp":4.68,"box":[0.0,0.03055555559694767,0.10208334028720856,0.1453703660517931],"occluded":false},{"timestamp":4.94,"box":[0.0,0.03055555559694767,0.08177083730697632,0.09907407499849796],"occluded":false},{"timestamp":5.2,"box":[0.0,0.03055555559694767,0.07135417312383652,0.09166666679084301],"occluded":false},{"timestamp":5.46,"box":[0.0,0.03055555559694767,0.05104167014360428,0.06481481529772282],"occluded":false},{"timestamp":8.84,"box":[0.17291668057441711,0.029629629105329514,0.18906251527369022,0.051851850003004074],"occluded":false}],"scores":[0.6717948913574219,0.6717948913574219,0.6717948913574219,0.6717948913574219,0.6717948913574219,0.6717948913574219,0.6717948913574219,0.6717948913574219,0.6717948913574219,0.6717948913574219,0.6717948913574219,0.6717948913574219,0.6717948913574219,0.6717948913574219,0.6717948913574219,0.6717948913574219,0.6717948913574219,0.6717948913574219,0.6717948913574219,0.6717948913574219]},{"noun_phrase":"woman with long dark hair","object_id":0,"observations":[{"timestamp":8.84,"box":[0.0,0.12314814329147339,0.16770833730697632,0.9731481075286865],"occluded":false},{"timestamp":9.1,"box":[0.0,0.05000000074505806,0.2880208492279053,0.9768518097698689],"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":[0.0,0.02777777798473835,0.2916666865348816,0.9759259093552828],"occluded":false},{"timestamp":10.66,"box":[0.0,0.03055555559694767,0.2666666805744171,0.9768518302589655],"occluded":false},{"timestamp":10.92,"box":[0.0,0.028703702613711357,0.2526041865348816,0.9749999772757292],"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":[0.8672199249267578,0.8672199249267578,0.8672199249267578,0.8672199249267578,0.8672199249267578,0.8672199249267578,0.8672199249267578,0.8672199249267578,0.8672199249267578,0.8672199249267578,0.8672199249267578,0.8672199249267578,0.8672199249267578,0.8672199249267578]}],"tracklets":[{"noun_phrase":"woman with long dark hair","object_id":0,"presentation_id":1,"observations":[{"timestamp":8.84,"box":[0.0,0.12314814329147339,0.16770833730697632,0.9731481075286865],"occluded":false},{"timestamp":9.1,"box":[0.0,0.05000000074505806,0.2880208492279053,0.9768518097698689],"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":[0.0,0.02777777798473835,0.2916666865348816,0.9759259093552828],"occluded":false},{"timestamp":10.66,"box":[0.0,0.03055555559694767,0.2666666805744171,0.9768518302589655],"occluded":false},{"timestamp":10.92,"box":[0.0,0.028703702613711357,0.2526041865348816,0.9749999772757292],"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":0.8672199249267578},{"noun_phrase":"glowing streetlamp","object_id":0,"presentation_id":2,"observations":[{"timestamp":0.78,"box":[0.06458333879709244,0.031481482088565826,0.13854167610406876,0.14537037163972855],"occluded":false},{"timestamp":1.04,"box":[0.15781250596046448,0.03055555559694767,0.22604167461395264,0.12592592276632786],"occluded":false},{"timestamp":1.3,"box":[0.1067708358168602,0.03055555559694767,0.18177083879709244,0.09166666679084301],"occluded":false},{"timestamp":1.56,"box":[0.0520833358168602,0.03055555559694767,0.11093750596046448,0.06574073992669582],"occluded":false},{"timestamp":1.82,"box":[0.02968750149011612,0.03055555559694767,0.12083333730697632,0.13611111231148243],"occluded":false},{"timestamp":2.08,"box":[0.05104167014360428,0.03055555559694767,0.14114584028720856,0.14351851679384708],"occluded":false},{"timestamp":2.34,"box":[0.0911458358168602,0.03055555559694767,0.17656251043081284,0.3240740690380335],"occluded":false},{"timestamp":2.6,"box":[0.12708334624767303,0.04722222313284874,0.20781251788139343,0.17777777835726738],"occluded":false},{"timestamp":2.86,"box":[0.10572917014360428,0.09074074029922485,0.19739583879709244,0.43518516421318054],"occluded":false},{"timestamp":3.12,"box":[0.07447917014360428,0.08425925672054291,0.17083334177732468,0.4287036806344986],"occluded":false},{"timestamp":3.38,"box":[0.08437500149011612,0.06203703582286835,0.17500000447034836,0.21111111342906952],"occluded":false},{"timestamp":3.64,"box":[0.10208334028720856,0.05092592537403107,0.20000001043081284,0.3453703671693802],"occluded":false},{"timestamp":3.9,"box":[0.09791667014360428,0.032407406717538834,0.18697917461395264,0.18703703954815865],"occluded":false},{"timestamp":4.16,"box":[0.07395833730697632,0.03333333134651184,0.16822917759418488,0.18518517911434174],"occluded":false},{"timestamp":4.42,"box":[0.04010416939854622,0.031481482088565826,0.1432291753590107,0.1787037029862404],"occluded":false},{"timestamp":4.68,"box":[0.0,0.03055555559694767,0.10208334028720856,0.1453703660517931],"occluded":false},{"timestamp":4.94,"box":[0.0,0.03055555559694767,0.08177083730697632,0.09907407499849796],"occluded":false},{"timestamp":5.2,"box":[0.0,0.03055555559694767,0.07135417312383652,0.09166666679084301],"occluded":false},{"timestamp":5.46,"box":[0.0,0.03055555559694767,0.05104167014360428,0.06481481529772282],"occluded":false},{"timestamp":8.84,"box":[0.17291668057441711,0.029629629105329514,0.18906251527369022,0.051851850003004074],"occluded":false}],"mean_score":0.6717948913574219}],"dropped":{},"fates":[{"noun_phrase":"glowing streetlamp","object_id":0,"frames":20,"mean_score":0.6717948913574219,"mean_area":0.012557943086135535,"outcome":"kept","presentation_id":2,"merged_into_phrase":null,"merged_into_object_id":null,"merge_iou":null},{"noun_phrase":"woman with long dark hair","object_id":0,"frames":14,"mean_score":0.8672199249267578,"mean_area":0.19698554652588612,"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/tmp/rmot_smoke_real_86260/renders/--cif7Nv6BI_chunk_014_scene_018_choices.mp4","media_bytes":3205106,"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","temperature":0.0,"as_video":true,"media_bytes":3205106,"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 woman with long dark hair who briefly appears on the left\n2. the glowing streetlamp mounted on the left wall\n3. the black car driving down the cobblestone street","questions":["the woman with long dark hair who briefly appears on the left","the glowing streetlamp mounted on the left wall","the black car driving down the cobblestone street"]},"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\": 2,\n      \"end_frame\": 6,\n      \"confidence\": 1.0\n    },\n    {\n      \"description_index\": 3,\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":10,"end_frame":13,"confidence":1.0},{"description_index":2,"matching_ids":[2],"start_frame":2,"end_frame":6,"confidence":1.0},{"description_index":3,"matching_ids":[],"start_frame":1,"end_frame":13,"confidence":1.0}],"answers":[{"expression":"the woman with long dark hair who briefly appears on the left","matching_presentation_ids":[1],"start_choice_frame":10,"end_choice_frame":13,"confidence":1.0,"self_reported_confidence":1.0},{"expression":"the glowing streetlamp mounted on the left wall","matching_presentation_ids":[2],"start_choice_frame":2,"end_choice_frame":6,"confidence":1.0,"self_reported_confidence":1.0},{"expression":"the black car driving down the cobblestone street","matching_presentation_ids":[],"start_choice_frame":1,"end_choice_frame":13,"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 woman with long dark hair who briefly appears on the left","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],[0.0,0.02777777798473835,0.2916666865348816,0.9759259093552828],[0.0,0.03055555559694767,0.2666666805744171,0.9768518302589655],[0.0,0.028703702613711357,0.2526041865348816,0.9749999772757292],[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 glowing streetlamp mounted on the left wall","kind":"spatial","is_negative":false,"tracks":[{"timestamps":[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":[[0.15781250596046448,0.03055555559694767,0.22604167461395264,0.12592592276632786],[0.1067708358168602,0.03055555559694767,0.18177083879709244,0.09166666679084301],[0.0520833358168602,0.03055555559694767,0.11093750596046448,0.06574073992669582],[0.02968750149011612,0.03055555559694767,0.12083333730697632,0.13611111231148243],[0.05104167014360428,0.03055555559694767,0.14114584028720856,0.14351851679384708],[0.0911458358168602,0.03055555559694767,0.17656251043081284,0.3240740690380335],[0.12708334624767303,0.04722222313284874,0.20781251788139343,0.17777777835726738],[0.10572917014360428,0.09074074029922485,0.19739583879709244,0.43518516421318054],[0.07447917014360428,0.08425925672054291,0.17083334177732468,0.4287036806344986],[0.08437500149011612,0.06203703582286835,0.17500000447034836,0.21111111342906952],[0.10208334028720856,0.05092592537403107,0.20000001043081284,0.3453703671693802],[0.09791667014360428,0.032407406717538834,0.18697917461395264,0.18703703954815865],[0.07395833730697632,0.03333333134651184,0.16822917759418488,0.18518517911434174],[0.04010416939854622,0.031481482088565826,0.1432291753590107,0.1787037029862404],[0.0,0.03055555559694767,0.10208334028720856,0.1453703660517931],[0.0,0.03055555559694767,0.08177083730697632,0.09907407499849796],[0.0,0.03055555559694767,0.07135417312383652,0.09166666679084301]]}]}],"rejected":{"the black car driving down 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_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 woman with long dark hair who briefly appears on the left","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],[0.0,0.02777777798473835,0.2916666865348816,0.9759259093552828],[0.0,0.03055555559694767,0.2666666805744171,0.9768518302589655],[0.0,0.028703702613711357,0.2526041865348816,0.9749999772757292],[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 glowing streetlamp mounted on the left wall","kind":"spatial","is_negative":false,"tracks":[{"timestamps":[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":[[0.15781250596046448,0.03055555559694767,0.22604167461395264,0.12592592276632786],[0.1067708358168602,0.03055555559694767,0.18177083879709244,0.09166666679084301],[0.0520833358168602,0.03055555559694767,0.11093750596046448,0.06574073992669582],[0.02968750149011612,0.03055555559694767,0.12083333730697632,0.13611111231148243],[0.05104167014360428,0.03055555559694767,0.14114584028720856,0.14351851679384708],[0.0911458358168602,0.03055555559694767,0.17656251043081284,0.3240740690380335],[0.12708334624767303,0.04722222313284874,0.20781251788139343,0.17777777835726738],[0.10572917014360428,0.09074074029922485,0.19739583879709244,0.43518516421318054],[0.07447917014360428,0.08425925672054291,0.17083334177732468,0.4287036806344986],[0.08437500149011612,0.06203703582286835,0.17500000447034836,0.21111111342906952],[0.10208334028720856,0.05092592537403107,0.20000001043081284,0.3453703671693802],[0.09791667014360428,0.032407406717538834,0.18697917461395264,0.18703703954815865],[0.07395833730697632,0.03333333134651184,0.16822917759418488,0.18518517911434174],[0.04010416939854622,0.031481482088565826,0.1432291753590107,0.1787037029862404],[0.0,0.03055555559694767,0.10208334028720856,0.1453703660517931],[0.0,0.03055555559694767,0.08177083730697632,0.09907407499849796],[0.0,0.03055555559694767,0.07135417312383652,0.09166666679084301]]}]}],"rmot_expression_count":2,"rmot_negative_count":0,"rmot_motion_count":0}}}}}]}