id int64 0 1.51k | level stringclasses 4
values | dataset stringclasses 2
values | gt_ans_text stringclasses 2
values | images images listlengths 2 2 | demo_prompt stringclasses 1
value | prompt_template stringclasses 1
value | qa_template dict | qa dict | metadata dict | depth_images images listlengths 2 2 | poses listlengths 2 2 | intrinsic listlengths 3 3 | relative_pose_vector dict | tau float64 15 65 | central_point_deviation float64 5.04 600 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
0 | 15-20 | ScanNet | Move left while yawing right | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | {
"type": [
"output_text",
"no_shuffle",
"thinking_tags",
"ans_tags"
],
"ans_candidates": "Move left while yawing right\nMove right while yawing left",
"output_format": "Provide your spatial reasoning inside <thinking></thinking> XML tags, and provide your answer inside <ans></ans> XML tags, e.g... | {
"w/ trap": {
"gt_idx": 2,
"gt_text": "Move left while yawing right",
"prompt": "<input>\nYou are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two f... | {
"scene_name": "scene0777_00",
"seq_name": null
} | [
[
[
-0.653784,
-0.246575,
0.71538,
5.198396
],
[
-0.753024,
0.11917,
-0.647112,
3.014966
],
[
0.07431,
-0.961769,
-0.263588,
1.481973
],
[
0,
0,
0,
1
]
],
[
[
-0.692992,
... | [
[
1165.484009,
0,
654.941589
],
[
0,
1164.543945,
477.277008
],
[
0,
0,
1
]
] | {
"text": [
"pitch up",
"yaw right",
"roll clockwise",
"translate left",
"translate up",
"translate forward"
],
"value": [
0.8695388261170335,
2.5237403635615676,
2.036576051847528,
-0.5300731706490858,
-0.2126121849777114,
0.5347742356362766
]
} | 15.312424 | 154.222516 | ||
1 | 15-20 | ScanNet | Move right while yawing left | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | {
"type": [
"output_text",
"no_shuffle",
"thinking_tags",
"ans_tags"
],
"ans_candidates": "Move left while yawing right\nMove right while yawing left",
"output_format": "Provide your spatial reasoning inside <thinking></thinking> XML tags, and provide your answer inside <ans></ans> XML tags, e.g... | {
"w/ trap": {
"gt_idx": 2,
"gt_text": "Move right while yawing left",
"prompt": "<input>\nYou are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two f... | {
"scene_name": "scene0720_00",
"seq_name": null
} | [
[
[
-0.850004,
0.324045,
-0.415316,
1.737525
],
[
0.525996,
0.564969,
-0.635718,
2.691831
],
[
0.028639,
-0.758818,
-0.650673,
1.437384
],
[
0,
0,
0,
1
]
],
[
[
-0.957391,
... | [
[
1165.371094,
0,
650.862061
],
[
0,
1166.109009,
488.595032
],
[
0,
0,
1
]
] | {
"text": [
"pitch down",
"yaw left",
"roll counterclockwise",
"translate right",
"translate up",
"translate forward"
],
"value": [
-0.2325136558727125,
-11.559676010194083,
-9.639876183129045,
0.2467825197134783,
-0.00931609492151575,
0.050318401252594125
]
} | 16.974416 | 178.478546 | ||
2 | 15-20 | ScanNet | Move right while yawing left | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | {
"type": [
"output_text",
"no_shuffle",
"thinking_tags",
"ans_tags"
],
"ans_candidates": "Move left while yawing right\nMove right while yawing left",
"output_format": "Provide your spatial reasoning inside <thinking></thinking> XML tags, and provide your answer inside <ans></ans> XML tags, e.g... | {
"w/ trap": {
"gt_idx": 1,
"gt_text": "Move right while yawing left",
"prompt": "<input>\nYou are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two f... | {
"scene_name": "scene0761_00",
"seq_name": null
} | [
[
[
0.110126,
0.164349,
-0.980235,
2.76086
],
[
0.992873,
-0.063385,
0.100919,
1.054285
],
[
-0.045547,
-0.984363,
-0.170159,
1.565021
],
[
0,
0,
0,
1
]
],
[
[
-0.229807,
... | [
[
1165.484009,
0,
654.941589
],
[
0,
1164.543945,
477.277008
],
[
0,
0,
1
]
] | {
"text": [
"pitch down",
"yaw left",
"roll counterclockwise",
"translate right",
"translate down",
"translate forward"
],
"value": [
-8.467240913410267,
-19.12631378577759,
-4.561333015383718,
0.8578359428758444,
0.04011273162058426,
0.13277380258229599
]
} | 18.365576 | 284.823045 | ||
3 | 15-20 | ScanNet | Move left while yawing right | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | {
"type": [
"output_text",
"no_shuffle",
"thinking_tags",
"ans_tags"
],
"ans_candidates": "Move left while yawing right\nMove right while yawing left",
"output_format": "Provide your spatial reasoning inside <thinking></thinking> XML tags, and provide your answer inside <ans></ans> XML tags, e.g... | {
"w/ trap": {
"gt_idx": 1,
"gt_text": "Move left while yawing right",
"prompt": "<input>\nYou are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two f... | {
"scene_name": "scene0780_00",
"seq_name": null
} | [
[
[
-0.610657,
0.020536,
-0.791628,
2.896583
],
[
0.7886,
0.106862,
-0.605549,
6.062631
],
[
0.072159,
-0.994062,
-0.081451,
1.528984
],
[
0,
0,
0,
1
]
],
[
[
-0.142475,
... | [
[
1165.484009,
0,
654.941589
],
[
0,
1164.543945,
477.277008
],
[
0,
0,
1
]
] | {
"text": [
"pitch down",
"yaw right",
"roll clockwise",
"translate left",
"translate up",
"translate backward"
],
"value": [
-0.31578882320992996,
29.13998921681101,
6.466588562614291,
-0.5965124390037193,
-0.03737335603049363,
-0.202739168870667
]
} | 15.100416 | 257.293141 | ||
4 | 15-20 | ScanNet | Move left while yawing right | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | {
"type": [
"output_text",
"no_shuffle",
"thinking_tags",
"ans_tags"
],
"ans_candidates": "Move left while yawing right\nMove right while yawing left",
"output_format": "Provide your spatial reasoning inside <thinking></thinking> XML tags, and provide your answer inside <ans></ans> XML tags, e.g... | {
"w/ trap": {
"gt_idx": 1,
"gt_text": "Move left while yawing right",
"prompt": "<input>\nYou are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two f... | {
"scene_name": "scene0736_00",
"seq_name": null
} | [
[
[
0.118582,
-0.490716,
0.863213,
1.472594
],
[
-0.99168,
-0.102379,
0.07803,
3.543334
],
[
0.050085,
-0.865284,
-0.498773,
1.651134
],
[
0,
0,
0,
1
]
],
[
[
-0.228461,
... | [
[
1163.445068,
0,
653.626038
],
[
0,
1164.793945,
481.600037
],
[
0,
0,
1
]
] | {
"text": [
"pitch up",
"yaw right",
"roll clockwise",
"translate left",
"translate up",
"translate forward"
],
"value": [
0.3113047836957608,
14.588166416993822,
14.880419262280965,
-0.3464036717700436,
-0.06554164290924014,
0.06192898782106482
]
} | 15.251165 | 56.958053 | ||
5 | 15-20 | ScanNet | Move left while yawing right | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | {
"type": [
"output_text",
"no_shuffle",
"thinking_tags",
"ans_tags"
],
"ans_candidates": "Move left while yawing right\nMove right while yawing left",
"output_format": "Provide your spatial reasoning inside <thinking></thinking> XML tags, and provide your answer inside <ans></ans> XML tags, e.g... | {
"w/ trap": {
"gt_idx": 0,
"gt_text": "Move left while yawing right",
"prompt": "<input>\nYou are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two f... | {
"scene_name": "scene0734_00",
"seq_name": null
} | [
[
[
0.561926,
-0.245396,
0.789949,
3.937602
],
[
-0.8269,
-0.191818,
0.528623,
1.583191
],
[
0.021804,
-0.950256,
-0.310706,
1.500637
],
[
0,
0,
0,
1
]
],
[
[
0.344536,
... | [
[
1165.723022,
0,
649.094971
],
[
0,
1165.738037,
484.765015
],
[
0,
0,
1
]
] | {
"text": [
"pitch up",
"yaw right",
"roll clockwise",
"translate left",
"translate down",
"translate backward"
],
"value": [
0.3960302423672985,
14.022424595951312,
2.1330166078381523,
-0.4903709993865239,
0.1580673270979709,
-0.6542405851494131
]
} | 17.363483 | 503.690353 | ||
6 | 15-20 | ScanNet | Move right while yawing left | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | {
"type": [
"output_text",
"no_shuffle",
"thinking_tags",
"ans_tags"
],
"ans_candidates": "Move left while yawing right\nMove right while yawing left",
"output_format": "Provide your spatial reasoning inside <thinking></thinking> XML tags, and provide your answer inside <ans></ans> XML tags, e.g... | {
"w/ trap": {
"gt_idx": 1,
"gt_text": "Move right while yawing left",
"prompt": "<input>\nYou are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two f... | {
"scene_name": "scene0733_00",
"seq_name": null
} | [
[
[
-0.978647,
0.079359,
-0.189609,
2.307856
],
[
0.203353,
0.508211,
-0.836881,
2.67656
],
[
0.029947,
-0.857569,
-0.513497,
1.513799
],
[
0,
0,
0,
1
]
],
[
[
-0.994915,
... | [
[
1165.371094,
0,
650.862061
],
[
0,
1166.109009,
488.595032
],
[
0,
0,
1
]
] | {
"text": [
"pitch down",
"yaw left",
"roll counterclockwise",
"translate right",
"translate up",
"translate forward"
],
"value": [
-1.8541919142390326,
-5.437159733640582,
-2.815387802932627,
0.4178683422939373,
-0.078982036272196,
0.1650631826682285
]
} | 15.489374 | 120.913869 | ||
7 | 15-20 | ScanNet | Move left while yawing right | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | {
"type": [
"output_text",
"no_shuffle",
"thinking_tags",
"ans_tags"
],
"ans_candidates": "Move left while yawing right\nMove right while yawing left",
"output_format": "Provide your spatial reasoning inside <thinking></thinking> XML tags, and provide your answer inside <ans></ans> XML tags, e.g... | {
"w/ trap": {
"gt_idx": 1,
"gt_text": "Move left while yawing right",
"prompt": "<input>\nYou are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two f... | {
"scene_name": "scene0795_00",
"seq_name": null
} | [
[
[
0.392256,
-0.398265,
0.829169,
1.308462
],
[
-0.919531,
-0.145813,
0.364968,
1.93661
],
[
-0.02445,
-0.905607,
-0.423412,
1.4516
],
[
0,
0,
0,
1
]
],
[
[
0.245726,
... | [
[
1165.484009,
0,
654.941589
],
[
0,
1164.543945,
477.277008
],
[
0,
0,
1
]
] | {
"text": [
"pitch down",
"yaw right",
"roll clockwise",
"translate left",
"translate down",
"translate backward"
],
"value": [
-0.4532305661363518,
8.54644229857336,
2.695496770345124,
-0.5217744989269781,
0.11009223970572002,
-0.23896327081994262
]
} | 16.002286 | 492.419362 | ||
8 | 15-20 | ScanNet | Move left while yawing right | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | {
"type": [
"output_text",
"no_shuffle",
"thinking_tags",
"ans_tags"
],
"ans_candidates": "Move left while yawing right\nMove right while yawing left",
"output_format": "Provide your spatial reasoning inside <thinking></thinking> XML tags, and provide your answer inside <ans></ans> XML tags, e.g... | {
"w/ trap": {
"gt_idx": 0,
"gt_text": "Move left while yawing right",
"prompt": "<input>\nYou are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two f... | {
"scene_name": "scene0754_00",
"seq_name": null
} | [
[
[
0.863925,
0.098248,
-0.493945,
1.924708
],
[
0.501118,
-0.07004,
0.86254,
2.791641
],
[
0.050146,
-0.992694,
-0.109743,
1.489921
],
[
0,
0,
0,
1
]
],
[
[
0.979708,
... | [
[
1165.723022,
0,
649.094971
],
[
0,
1165.738037,
484.765015
],
[
0,
0,
1
]
] | {
"text": [
"pitch down",
"yaw right",
"roll clockwise",
"translate left",
"translate up",
"translate forward"
],
"value": [
-0.28090306956047384,
18.038855291599944,
6.341010257376063,
-0.3299501307310373,
-0.09147576654313494,
0.19781457894473853
]
} | 16.533395 | 87.189318 | ||
9 | 15-20 | ScanNet | Move left while yawing right | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | {
"type": [
"output_text",
"no_shuffle",
"thinking_tags",
"ans_tags"
],
"ans_candidates": "Move left while yawing right\nMove right while yawing left",
"output_format": "Provide your spatial reasoning inside <thinking></thinking> XML tags, and provide your answer inside <ans></ans> XML tags, e.g... | {
"w/ trap": {
"gt_idx": 1,
"gt_text": "Move left while yawing right",
"prompt": "<input>\nYou are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two f... | {
"scene_name": "scene0753_00",
"seq_name": null
} | [
[
[
-0.65427,
-0.288554,
0.699048,
2.661347
],
[
-0.755437,
0.206232,
-0.62192,
2.934681
],
[
0.035291,
-0.93499,
-0.352915,
1.427871
],
[
0,
0,
0,
1
]
],
[
[
-0.908781,
... | [
[
1165.723022,
0,
649.094971
],
[
0,
1165.738037,
484.765015
],
[
0,
0,
1
]
] | {
"text": [
"pitch down",
"yaw right",
"roll clockwise",
"translate left",
"translate up",
"translate forward"
],
"value": [
-0.03171839077610357,
23.353947901265112,
7.99361359295216,
-0.31482354458687967,
-0.07706709336605044,
0.23056125104999178
]
} | 15.349827 | 167.999203 | ||
10 | 15-20 | ScanNet | Move left while yawing right | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | {
"type": [
"output_text",
"no_shuffle",
"thinking_tags",
"ans_tags"
],
"ans_candidates": "Move left while yawing right\nMove right while yawing left",
"output_format": "Provide your spatial reasoning inside <thinking></thinking> XML tags, and provide your answer inside <ans></ans> XML tags, e.g... | {
"w/ trap": {
"gt_idx": 0,
"gt_text": "Move left while yawing right",
"prompt": "<input>\nYou are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two f... | {
"scene_name": "scene0763_00",
"seq_name": null
} | [
[
[
-0.845212,
-0.169936,
-0.506693,
1.373511
],
[
0.520027,
-0.04288,
-0.853072,
2.047248
],
[
0.123241,
-0.984522,
0.124615,
1.893714
],
[
0,
0,
0,
1
]
],
[
[
-0.624619,
... | [
[
1165.371094,
0,
650.862061
],
[
0,
1166.109009,
488.595032
],
[
0,
0,
1
]
] | {
"text": [
"pitch down",
"yaw right",
"roll counterclockwise",
"translate left",
"translate down",
"translate forward"
],
"value": [
-7.873898377320575,
19.25371113737093,
-2.045754719320441,
-0.3411987536134592,
0.1626043252249647,
0.7174915625984759
]
} | 15.601222 | 147.313457 | ||
11 | 15-20 | ScanNet | Move right while yawing left | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | {
"type": [
"output_text",
"no_shuffle",
"thinking_tags",
"ans_tags"
],
"ans_candidates": "Move left while yawing right\nMove right while yawing left",
"output_format": "Provide your spatial reasoning inside <thinking></thinking> XML tags, and provide your answer inside <ans></ans> XML tags, e.g... | {
"w/ trap": {
"gt_idx": 0,
"gt_text": "Move right while yawing left",
"prompt": "<input>\nYou are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two f... | {
"scene_name": "scene0741_00",
"seq_name": null
} | [
[
[
0.003083,
-0.456645,
0.889644,
1.339768
],
[
-0.99996,
0.006093,
0.006593,
2.135948
],
[
-0.008431,
-0.889628,
-0.456608,
1.443262
],
[
0,
0,
0,
1
]
],
[
[
0.481506,
... | [
[
1165.723022,
0,
649.094971
],
[
0,
1165.738037,
484.765015
],
[
0,
0,
1
]
] | {
"text": [
"pitch up",
"yaw left",
"roll counterclockwise",
"translate right",
"translate up",
"translate forward"
],
"value": [
3.548217589792395,
-22.895977136355405,
-18.38614816878349,
0.6294860162981089,
-0.011944577051586025,
0.03937739502995752
]
} | 15.689701 | 200.820448 | ||
12 | 15-20 | ScanNet | Move left while yawing right | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | {
"type": [
"output_text",
"no_shuffle",
"thinking_tags",
"ans_tags"
],
"ans_candidates": "Move left while yawing right\nMove right while yawing left",
"output_format": "Provide your spatial reasoning inside <thinking></thinking> XML tags, and provide your answer inside <ans></ans> XML tags, e.g... | {
"w/ trap": {
"gt_idx": 2,
"gt_text": "Move left while yawing right",
"prompt": "<input>\nYou are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two f... | {
"scene_name": "scene0747_00",
"seq_name": null
} | [
[
[
0.946909,
-0.091652,
0.30816,
5.03169
],
[
-0.308341,
-0.530321,
0.789737,
2.087131
],
[
0.091042,
-0.842828,
-0.530426,
1.463881
],
[
0,
0,
0,
1
]
],
[
[
0.776531,
... | [
[
1161.745972,
0,
658.041992
],
[
0,
1159.107056,
486.467102
],
[
0,
0,
1
]
] | {
"text": [
"pitch down",
"yaw right",
"roll clockwise",
"translate left",
"translate up",
"translate forward"
],
"value": [
-1.7053328861546142,
15.693081972670877,
14.609166053361166,
-0.5132226177031733,
-0.171202601212324,
0.25096720826967056
]
} | 15.856791 | 53.948846 | ||
13 | 15-20 | ScanNet | Move left while yawing right | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | {
"type": [
"output_text",
"no_shuffle",
"thinking_tags",
"ans_tags"
],
"ans_candidates": "Move left while yawing right\nMove right while yawing left",
"output_format": "Provide your spatial reasoning inside <thinking></thinking> XML tags, and provide your answer inside <ans></ans> XML tags, e.g... | {
"w/ trap": {
"gt_idx": 0,
"gt_text": "Move left while yawing right",
"prompt": "<input>\nYou are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two f... | {
"scene_name": "scene0725_00",
"seq_name": null
} | [
[
[
-0.285922,
0.288589,
-0.913765,
1.183743
],
[
0.957933,
0.061442,
-0.280338,
1.285102
],
[
-0.024759,
-0.95548,
-0.294016,
1.706392
],
[
0,
0,
0,
1
]
],
[
[
0.365249,
... | [
[
1165.371094,
0,
650.862061
],
[
0,
1166.109009,
488.595032
],
[
0,
0,
1
]
] | {
"text": [
"pitch up",
"yaw right",
"roll clockwise",
"translate left",
"translate up",
"translate forward"
],
"value": [
4.2150399253018795,
35.4590251238866,
14.854862457428252,
-0.1777950476382989,
-0.08776298440920471,
0.07587955732947727
]
} | 19.691117 | 336.809742 | ||
14 | 15-20 | ScanNet | Move left while yawing right | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | {
"type": [
"output_text",
"no_shuffle",
"thinking_tags",
"ans_tags"
],
"ans_candidates": "Move left while yawing right\nMove right while yawing left",
"output_format": "Provide your spatial reasoning inside <thinking></thinking> XML tags, and provide your answer inside <ans></ans> XML tags, e.g... | {
"w/ trap": {
"gt_idx": 2,
"gt_text": "Move left while yawing right",
"prompt": "<input>\nYou are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two f... | {
"scene_name": "scene0794_00",
"seq_name": null
} | [
[
[
-0.617309,
-0.437666,
0.653741,
1.775739
],
[
-0.786689,
0.350855,
-0.507958,
4.534101
],
[
-0.007053,
-0.827859,
-0.560893,
1.321378
],
[
0,
0,
0,
1
]
],
[
[
-0.898317,
... | [
[
1165.484009,
0,
654.941589
],
[
0,
1164.543945,
477.277008
],
[
0,
0,
1
]
] | {
"text": [
"pitch up",
"yaw right",
"roll clockwise",
"translate left",
"translate up",
"translate forward"
],
"value": [
6.308449960794756,
21.867897163291993,
14.15388741858674,
-0.5598895242225366,
-0.11289565201650886,
0.14209231064940542
]
} | 17.687465 | 175.845553 | ||
15 | 15-20 | ScanNet | Move right while yawing left | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | {
"type": [
"output_text",
"no_shuffle",
"thinking_tags",
"ans_tags"
],
"ans_candidates": "Move left while yawing right\nMove right while yawing left",
"output_format": "Provide your spatial reasoning inside <thinking></thinking> XML tags, and provide your answer inside <ans></ans> XML tags, e.g... | {
"w/ trap": {
"gt_idx": 0,
"gt_text": "Move right while yawing left",
"prompt": "<input>\nYou are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two f... | {
"scene_name": "scene0765_00",
"seq_name": null
} | [
[
[
0.749137,
-0.055233,
0.660108,
1.644927
],
[
-0.659482,
0.031474,
0.75106,
2.141366
],
[
-0.062259,
-0.997977,
-0.012847,
1.73127
],
[
0,
0,
0,
1
]
],
[
[
0.982649,
... | [
[
1165.371094,
0,
650.862061
],
[
0,
1166.109009,
488.595032
],
[
0,
0,
1
]
] | {
"text": [
"pitch up",
"yaw left",
"roll clockwise",
"translate right",
"translate up",
"translate forward"
],
"value": [
5.762236079429669,
-32.87912827429679,
4.0461034954515664,
0.15804644295567627,
-0.07233540800062799,
0.22520899651786497
]
} | 17.647331 | 400.849765 | ||
16 | 15-20 | ScanNet | Move right while yawing left | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | {
"type": [
"output_text",
"no_shuffle",
"thinking_tags",
"ans_tags"
],
"ans_candidates": "Move left while yawing right\nMove right while yawing left",
"output_format": "Provide your spatial reasoning inside <thinking></thinking> XML tags, and provide your answer inside <ans></ans> XML tags, e.g... | {
"w/ trap": {
"gt_idx": 1,
"gt_text": "Move right while yawing left",
"prompt": "<input>\nYou are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two f... | {
"scene_name": "scene0747_00",
"seq_name": null
} | [
[
[
0.565778,
-0.312571,
0.763016,
5.921611
],
[
-0.822471,
-0.279717,
0.495278,
5.855114
],
[
0.058619,
-0.907776,
-0.415339,
1.457494
],
[
0,
0,
0,
1
]
],
[
[
0.739641,
... | [
[
1161.745972,
0,
658.041992
],
[
0,
1159.107056,
486.467102
],
[
0,
0,
1
]
] | {
"text": [
"pitch down",
"yaw left",
"roll counterclockwise",
"translate right",
"translate up",
"translate forward"
],
"value": [
-3.1309001576286324,
-12.208091513343069,
-5.176430509510646,
0.4241887694376735,
-0.0045770625646150975,
0.12339671695247123
]
} | 15.064537 | 143.068638 | ||
17 | 15-20 | ScanNet | Move left while yawing right | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | {
"type": [
"output_text",
"no_shuffle",
"thinking_tags",
"ans_tags"
],
"ans_candidates": "Move left while yawing right\nMove right while yawing left",
"output_format": "Provide your spatial reasoning inside <thinking></thinking> XML tags, and provide your answer inside <ans></ans> XML tags, e.g... | {
"w/ trap": {
"gt_idx": 0,
"gt_text": "Move left while yawing right",
"prompt": "<input>\nYou are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two f... | {
"scene_name": "scene0775_00",
"seq_name": null
} | [
[
[
0.981958,
0.023483,
0.187637,
2.115612
],
[
-0.175258,
-0.259641,
0.949669,
2.117794
],
[
0.07102,
-0.96542,
-0.250841,
1.483329
],
[
0,
0,
0,
1
]
],
[
[
0.867068,
... | [
[
1165.484009,
0,
654.941589
],
[
0,
1164.543945,
477.277008
],
[
0,
0,
1
]
] | {
"text": [
"pitch down",
"yaw right",
"roll clockwise",
"translate left",
"translate up",
"translate forward"
],
"value": [
-5.176428438269492,
18.22507891435919,
8.499331296274027,
-0.26571594505749907,
-0.007076362279868409,
0.05481366783837904
]
} | 16.342044 | 110.867048 | ||
18 | 15-20 | ScanNet | Move left while yawing right | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | {
"type": [
"output_text",
"no_shuffle",
"thinking_tags",
"ans_tags"
],
"ans_candidates": "Move left while yawing right\nMove right while yawing left",
"output_format": "Provide your spatial reasoning inside <thinking></thinking> XML tags, and provide your answer inside <ans></ans> XML tags, e.g... | {
"w/ trap": {
"gt_idx": 1,
"gt_text": "Move left while yawing right",
"prompt": "<input>\nYou are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two f... | {
"scene_name": "scene0713_00",
"seq_name": null
} | [
[
[
0.918968,
-0.227759,
0.321906,
1.422764
],
[
-0.386886,
-0.678641,
0.624312,
1.303139
],
[
0.076266,
-0.698263,
-0.711767,
1.275649
],
[
0,
0,
0,
1
]
],
[
[
0.718335,
... | [
[
1161.745972,
0,
658.041992
],
[
0,
1159.107056,
486.467102
],
[
0,
0,
1
]
] | {
"text": [
"pitch up",
"yaw right",
"roll clockwise",
"translate left",
"translate up",
"translate forward"
],
"value": [
1.9695514335198974,
14.792413661677331,
15.092751864139116,
-0.39869172628754956,
-0.04234738270147753,
0.10742558630877153
]
} | 15.432304 | 63.0777 | ||
19 | 15-20 | ScanNet | Move right while yawing left | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | {
"type": [
"output_text",
"no_shuffle",
"thinking_tags",
"ans_tags"
],
"ans_candidates": "Move left while yawing right\nMove right while yawing left",
"output_format": "Provide your spatial reasoning inside <thinking></thinking> XML tags, and provide your answer inside <ans></ans> XML tags, e.g... | {
"w/ trap": {
"gt_idx": 0,
"gt_text": "Move right while yawing left",
"prompt": "<input>\nYou are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two f... | {
"scene_name": "scene0770_00",
"seq_name": null
} | [
[
[
0.611773,
0.329869,
-0.718971,
7.733308
],
[
0.790647,
-0.283381,
0.542745,
4.052631
],
[
-0.024708,
-0.90049,
-0.434175,
1.422027
],
[
0,
0,
0,
1
]
],
[
[
0.115536,
... | [
[
1165.484009,
0,
654.941589
],
[
0,
1164.543945,
477.277008
],
[
0,
0,
1
]
] | {
"text": [
"pitch up",
"yaw left",
"roll counterclockwise",
"translate right",
"translate up",
"translate forward"
],
"value": [
1.0047268249566579,
-26.43578480876271,
-17.243912245309854,
0.19882422377622078,
-0.09674394634900807,
0.26927715167099775
]
} | 16.595818 | 368.845564 | ||
20 | 15-20 | ScanNet | Move left while yawing right | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | {
"type": [
"output_text",
"no_shuffle",
"thinking_tags",
"ans_tags"
],
"ans_candidates": "Move left while yawing right\nMove right while yawing left",
"output_format": "Provide your spatial reasoning inside <thinking></thinking> XML tags, and provide your answer inside <ans></ans> XML tags, e.g... | {
"w/ trap": {
"gt_idx": 0,
"gt_text": "Move left while yawing right",
"prompt": "<input>\nYou are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two f... | {
"scene_name": "scene0781_00",
"seq_name": null
} | [
[
[
0.961202,
-0.075615,
0.26528,
3.842405
],
[
-0.27372,
-0.380647,
0.883281,
1.288019
],
[
0.034189,
-0.921624,
-0.386576,
1.468105
],
[
0,
0,
0,
1
]
],
[
[
0.89664,
... | [
[
1165.484009,
0,
654.941589
],
[
0,
1164.543945,
477.277008
],
[
0,
0,
1
]
] | {
"text": [
"pitch up",
"yaw right",
"roll clockwise",
"translate left",
"translate up",
"translate forward"
],
"value": [
0.4142721352039328,
10.383837352246646,
0.7169682115307298,
-0.4193263405645342,
-0.07279038611172872,
0.20428329371785892
]
} | 15.721616 | 144.301065 | ||
21 | 15-20 | ScanNet | Move right while yawing left | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | {
"type": [
"output_text",
"no_shuffle",
"thinking_tags",
"ans_tags"
],
"ans_candidates": "Move left while yawing right\nMove right while yawing left",
"output_format": "Provide your spatial reasoning inside <thinking></thinking> XML tags, and provide your answer inside <ans></ans> XML tags, e.g... | {
"w/ trap": {
"gt_idx": 1,
"gt_text": "Move right while yawing left",
"prompt": "<input>\nYou are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two f... | {
"scene_name": "scene0781_00",
"seq_name": null
} | [
[
[
-0.59312,
-0.314972,
0.740946,
2.092771
],
[
-0.805022,
0.245902,
-0.539881,
4.001449
],
[
-0.012153,
-0.916692,
-0.399409,
1.402681
],
[
0,
0,
0,
1
]
],
[
[
-0.544047,
... | [
[
1165.484009,
0,
654.941589
],
[
0,
1164.543945,
477.277008
],
[
0,
0,
1
]
] | {
"text": [
"pitch down",
"yaw left",
"roll clockwise",
"translate right",
"translate up",
"translate forward"
],
"value": [
-0.17404881042891696,
-3.9786129490245683,
0.6946929505613065,
0.31997347169089174,
-0.15698693978719536,
0.325060362200011
]
} | 15.276519 | 180.615411 | ||
22 | 15-20 | ScanNet | Move left while yawing right | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | {
"type": [
"output_text",
"no_shuffle",
"thinking_tags",
"ans_tags"
],
"ans_candidates": "Move left while yawing right\nMove right while yawing left",
"output_format": "Provide your spatial reasoning inside <thinking></thinking> XML tags, and provide your answer inside <ans></ans> XML tags, e.g... | {
"w/ trap": {
"gt_idx": 1,
"gt_text": "Move left while yawing right",
"prompt": "<input>\nYou are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two f... | {
"scene_name": "scene0768_00",
"seq_name": null
} | [
[
[
-0.620533,
0.544869,
-0.563965,
1.454437
],
[
0.784149,
0.437507,
-0.44011,
4.271794
],
[
0.006936,
-0.715336,
-0.698746,
1.63638
],
[
0,
0,
0,
1
]
],
[
[
-0.163838,
... | [
[
1165.484009,
0,
654.941589
],
[
0,
1164.543945,
477.277008
],
[
0,
0,
1
]
] | {
"text": [
"pitch down",
"yaw right",
"roll clockwise",
"translate left",
"translate down",
"translate backward"
],
"value": [
-0.8737071152161131,
18.731806059450637,
22.542004260813467,
-0.34502567767372927,
0.08956839434103325,
-0.040550636475460866
]
} | 15.504571 | 186.581165 | ||
23 | 15-20 | ScanNet | Move left while yawing right | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | {
"type": [
"output_text",
"no_shuffle",
"thinking_tags",
"ans_tags"
],
"ans_candidates": "Move left while yawing right\nMove right while yawing left",
"output_format": "Provide your spatial reasoning inside <thinking></thinking> XML tags, and provide your answer inside <ans></ans> XML tags, e.g... | {
"w/ trap": {
"gt_idx": 2,
"gt_text": "Move left while yawing right",
"prompt": "<input>\nYou are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two f... | {
"scene_name": "scene0771_00",
"seq_name": null
} | [
[
[
-0.264498,
-0.360081,
0.894641,
2.429003
],
[
-0.962235,
0.036615,
-0.269745,
3.38669
],
[
0.064372,
-0.932202,
-0.356167,
1.45668
],
[
0,
0,
0,
1
]
],
[
[
-0.610637,
... | [
[
1165.484009,
0,
654.941589
],
[
0,
1164.543945,
477.277008
],
[
0,
0,
1
]
] | {
"text": [
"pitch down",
"yaw right",
"roll clockwise",
"translate left",
"translate up",
"translate forward"
],
"value": [
-2.198000013802051,
20.08558567186714,
10.006797937422983,
-0.33113452317732284,
-0.07223656097822495,
0.16464544747562948
]
} | 16.141785 | 51.301936 | ||
24 | 15-20 | ScanNet | Move right while yawing left | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | {
"type": [
"output_text",
"no_shuffle",
"thinking_tags",
"ans_tags"
],
"ans_candidates": "Move left while yawing right\nMove right while yawing left",
"output_format": "Provide your spatial reasoning inside <thinking></thinking> XML tags, and provide your answer inside <ans></ans> XML tags, e.g... | {
"w/ trap": {
"gt_idx": 1,
"gt_text": "Move right while yawing left",
"prompt": "<input>\nYou are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two f... | {
"scene_name": "scene0787_00",
"seq_name": null
} | [
[
[
-0.939263,
0.217903,
-0.265147,
3.241554
],
[
0.343076,
0.575635,
-0.742255,
2.983825
],
[
-0.009111,
-0.788139,
-0.61543,
1.479304
],
[
0,
0,
0,
1
]
],
[
[
-0.998275,
... | [
[
1165.371094,
0,
650.862061
],
[
0,
1166.109009,
488.595032
],
[
0,
0,
1
]
] | {
"text": [
"pitch up",
"yaw left",
"roll counterclockwise",
"translate right",
"translate up",
"translate forward"
],
"value": [
2.067996907783607,
-12.489237499154035,
-11.2775867330131,
0.34314527655008065,
-0.02447964381517731,
0.011824196711570423
]
} | 16.270924 | 36.968809 | ||
25 | 15-20 | ScanNet | Move right while yawing left | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | {
"type": [
"output_text",
"no_shuffle",
"thinking_tags",
"ans_tags"
],
"ans_candidates": "Move left while yawing right\nMove right while yawing left",
"output_format": "Provide your spatial reasoning inside <thinking></thinking> XML tags, and provide your answer inside <ans></ans> XML tags, e.g... | {
"w/ trap": {
"gt_idx": 1,
"gt_text": "Move right while yawing left",
"prompt": "<input>\nYou are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two f... | {
"scene_name": "scene0711_00",
"seq_name": null
} | [
[
[
0.997494,
-0.065028,
0.02789,
4.097476
],
[
-0.063227,
-0.642238,
0.763893,
1.700076
],
[
-0.031762,
-0.763742,
-0.64474,
1.212525
],
[
0,
0,
0,
1
]
],
[
[
0.891392,
... | [
[
1163.445068,
0,
653.626038
],
[
0,
1164.793945,
481.600037
],
[
0,
0,
1
]
] | {
"text": [
"pitch down",
"yaw left",
"roll counterclockwise",
"translate right",
"translate up",
"translate backward"
],
"value": [
-0.1229576919443332,
-21.703319077105846,
-22.16705958436291,
0.46438534096124817,
-0.0034183940956142855,
-0.032946670062720784
]
... | 15.028487 | 275.907872 | ||
26 | 15-20 | ScanNet | Move right while yawing left | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | {
"type": [
"output_text",
"no_shuffle",
"thinking_tags",
"ans_tags"
],
"ans_candidates": "Move left while yawing right\nMove right while yawing left",
"output_format": "Provide your spatial reasoning inside <thinking></thinking> XML tags, and provide your answer inside <ans></ans> XML tags, e.g... | {
"w/ trap": {
"gt_idx": 1,
"gt_text": "Move right while yawing left",
"prompt": "<input>\nYou are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two f... | {
"scene_name": "scene0777_00",
"seq_name": null
} | [
[
[
0.53836,
-0.209993,
0.816133,
8.017328
],
[
-0.839208,
-0.221859,
0.496497,
2.047173
],
[
0.076806,
-0.952198,
-0.295667,
1.525158
],
[
0,
0,
0,
1
]
],
[
[
0.667511,
... | [
[
1165.484009,
0,
654.941589
],
[
0,
1164.543945,
477.277008
],
[
0,
0,
1
]
] | {
"text": [
"pitch down",
"yaw left",
"roll counterclockwise",
"translate right",
"translate up",
"translate forward"
],
"value": [
-0.23237719093065604,
-9.211983415089097,
-1.55867019235645,
0.37069040416708576,
-0.023957394475530247,
0.19419289365153514
]
} | 15.288473 | 114.146236 | ||
27 | 15-20 | ScanNet | Move left while yawing right | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | {
"type": [
"output_text",
"no_shuffle",
"thinking_tags",
"ans_tags"
],
"ans_candidates": "Move left while yawing right\nMove right while yawing left",
"output_format": "Provide your spatial reasoning inside <thinking></thinking> XML tags, and provide your answer inside <ans></ans> XML tags, e.g... | {
"w/ trap": {
"gt_idx": 2,
"gt_text": "Move left while yawing right",
"prompt": "<input>\nYou are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two f... | {
"scene_name": "scene0747_00",
"seq_name": null
} | [
[
[
0.693412,
0.375355,
-0.615051,
6.2785
],
[
0.720198,
-0.334718,
0.607684,
7.511715
],
[
0.022228,
-0.864334,
-0.502427,
1.384494
],
[
0,
0,
0,
1
]
],
[
[
0.846632,
... | [
[
1161.745972,
0,
658.041992
],
[
0,
1159.107056,
486.467102
],
[
0,
0,
1
]
] | {
"text": [
"pitch up",
"yaw right",
"roll clockwise",
"translate left",
"translate up",
"translate forward"
],
"value": [
1.5819205181619955,
16.17905919628199,
1.5159473429986399,
-0.4433281104309241,
-0.17175553011112377,
0.1665970322542562
]
} | 15.038555 | 172.463054 | ||
28 | 15-20 | ScanNet | Move left while yawing right | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | {
"type": [
"output_text",
"no_shuffle",
"thinking_tags",
"ans_tags"
],
"ans_candidates": "Move left while yawing right\nMove right while yawing left",
"output_format": "Provide your spatial reasoning inside <thinking></thinking> XML tags, and provide your answer inside <ans></ans> XML tags, e.g... | {
"w/ trap": {
"gt_idx": 0,
"gt_text": "Move left while yawing right",
"prompt": "<input>\nYou are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two f... | {
"scene_name": "scene0723_00",
"seq_name": null
} | [
[
[
0.924176,
0.074056,
-0.374718,
2.914668
],
[
0.379907,
-0.076481,
0.921858,
2.471717
],
[
0.03961,
-0.994317,
-0.098816,
1.879926
],
[
0,
0,
0,
1
]
],
[
[
0.946588,
... | [
[
1165.371094,
0,
650.862061
],
[
0,
1166.109009,
488.595032
],
[
0,
0,
1
]
] | {
"text": [
"pitch down",
"yaw right",
"roll clockwise",
"translate left",
"translate down",
"translate forward"
],
"value": [
-0.07072612978213332,
3.361213661760218,
2.0185114874515278,
-0.15075033845220487,
0.012898053851727376,
0.08477397265847442
]
} | 15.666548 | 198.661206 | ||
29 | 15-20 | ScanNet | Move left while yawing right | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | {
"type": [
"output_text",
"no_shuffle",
"thinking_tags",
"ans_tags"
],
"ans_candidates": "Move left while yawing right\nMove right while yawing left",
"output_format": "Provide your spatial reasoning inside <thinking></thinking> XML tags, and provide your answer inside <ans></ans> XML tags, e.g... | {
"w/ trap": {
"gt_idx": 1,
"gt_text": "Move left while yawing right",
"prompt": "<input>\nYou are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two f... | {
"scene_name": "scene0799_00",
"seq_name": null
} | [
[
[
-0.919727,
-0.19387,
0.341347,
2.289529
],
[
-0.391617,
0.392905,
-0.832023,
4.09216
],
[
0.027188,
-0.898911,
-0.437288,
1.531799
],
[
0,
0,
0,
1
]
],
[
[
-0.979756,
... | [
[
1165.484009,
0,
654.941589
],
[
0,
1164.543945,
477.277008
],
[
0,
0,
1
]
] | {
"text": [
"pitch up",
"yaw right",
"roll clockwise",
"translate left",
"translate up",
"translate forward"
],
"value": [
4.670010997405423,
11.178183262685994,
4.106655853018674,
-0.4438701238349778,
-0.1210273073645258,
0.24975493716618757
]
} | 15.959302 | 215.979139 | ||
30 | 15-20 | ScanNet | Move right while yawing left | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | {
"type": [
"output_text",
"no_shuffle",
"thinking_tags",
"ans_tags"
],
"ans_candidates": "Move left while yawing right\nMove right while yawing left",
"output_format": "Provide your spatial reasoning inside <thinking></thinking> XML tags, and provide your answer inside <ans></ans> XML tags, e.g... | {
"w/ trap": {
"gt_idx": 0,
"gt_text": "Move right while yawing left",
"prompt": "<input>\nYou are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two f... | {
"scene_name": "scene0802_00",
"seq_name": null
} | [
[
[
0.391076,
0.66206,
-0.639326,
2.964664
],
[
0.92013,
-0.265752,
0.287642,
2.213233
],
[
0.020534,
-0.700752,
-0.71311,
2.043929
],
[
0,
0,
0,
1
]
],
[
[
0.168274,
0... | [
[
1165.371094,
0,
650.862061
],
[
0,
1166.109009,
488.595032
],
[
0,
0,
1
]
] | {
"text": [
"pitch down",
"yaw left",
"roll counterclockwise",
"translate right",
"translate up",
"translate forward"
],
"value": [
-0.04942734326629382,
-10.354582481160339,
-8.584194588109652,
0.3251446354944676,
-0.1012329890895893,
0.1641790740779836
]
} | 15.259907 | 100.703096 | ||
31 | 15-20 | ScanNet | Move right while yawing left | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | {
"type": [
"output_text",
"no_shuffle",
"thinking_tags",
"ans_tags"
],
"ans_candidates": "Move left while yawing right\nMove right while yawing left",
"output_format": "Provide your spatial reasoning inside <thinking></thinking> XML tags, and provide your answer inside <ans></ans> XML tags, e.g... | {
"w/ trap": {
"gt_idx": 1,
"gt_text": "Move right while yawing left",
"prompt": "<input>\nYou are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two f... | {
"scene_name": "scene0733_00",
"seq_name": null
} | [
[
[
-0.224427,
-0.39858,
0.88925,
2.480331
],
[
-0.974468,
0.097977,
-0.202019,
3.542562
],
[
-0.006606,
-0.911885,
-0.410392,
1.648452
],
[
0,
0,
0,
1
]
],
[
[
0.023187,
... | [
[
1165.371094,
0,
650.862061
],
[
0,
1166.109009,
488.595032
],
[
0,
0,
1
]
] | {
"text": [
"pitch down",
"yaw left",
"roll counterclockwise",
"translate right",
"translate down",
"translate backward"
],
"value": [
-5.964030238609337,
-11.936950625377644,
-8.332882579304304,
0.513976526225302,
0.12734153680424942,
-0.12337025629395149
]
} | 17.577905 | 365.03077 | ||
32 | 15-20 | ScanNet | Move left while yawing right | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | {
"type": [
"output_text",
"no_shuffle",
"thinking_tags",
"ans_tags"
],
"ans_candidates": "Move left while yawing right\nMove right while yawing left",
"output_format": "Provide your spatial reasoning inside <thinking></thinking> XML tags, and provide your answer inside <ans></ans> XML tags, e.g... | {
"w/ trap": {
"gt_idx": 2,
"gt_text": "Move left while yawing right",
"prompt": "<input>\nYou are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two f... | {
"scene_name": "scene0761_00",
"seq_name": null
} | [
[
[
-0.907419,
0.076123,
-0.413274,
1.995378
],
[
0.413494,
0.337068,
-0.845818,
2.496061
],
[
0.074916,
-0.938398,
-0.337338,
1.415632
],
[
0,
0,
0,
1
]
],
[
[
-0.474316,
... | [
[
1165.484009,
0,
654.941589
],
[
0,
1164.543945,
477.277008
],
[
0,
0,
1
]
] | {
"text": [
"pitch down",
"yaw right",
"roll clockwise",
"translate left",
"translate down",
"translate forward"
],
"value": [
-9.988030755199052,
34.23205425954442,
15.34973827091338,
-0.46536900362229483,
0.022269683023564302,
0.05403657374705917
]
} | 17.453563 | 432.377781 | ||
33 | 15-20 | ScanNet | Move right while yawing left | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | {
"type": [
"output_text",
"no_shuffle",
"thinking_tags",
"ans_tags"
],
"ans_candidates": "Move left while yawing right\nMove right while yawing left",
"output_format": "Provide your spatial reasoning inside <thinking></thinking> XML tags, and provide your answer inside <ans></ans> XML tags, e.g... | {
"w/ trap": {
"gt_idx": 0,
"gt_text": "Move right while yawing left",
"prompt": "<input>\nYou are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two f... | {
"scene_name": "scene0736_00",
"seq_name": null
} | [
[
[
0.235194,
-0.824879,
0.51406,
1.449278
],
[
-0.971622,
-0.21325,
0.102351,
3.00377
],
[
0.025196,
-0.523545,
-0.851626,
1.586954
],
[
0,
0,
0,
1
]
],
[
[
0.504416,
... | [
[
1163.445068,
0,
653.626038
],
[
0,
1164.793945,
481.600037
],
[
0,
0,
1
]
] | {
"text": [
"pitch down",
"yaw left",
"roll counterclockwise",
"translate right",
"translate up",
"translate forward"
],
"value": [
-1.2324430042093293,
-6.604123685565737,
-15.593185806830418,
0.21354645893401747,
-0.06566187938054657,
0.08279393210116859
]
} | 15.893086 | 236.260548 | ||
34 | 15-20 | ScanNet | Move right while yawing left | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | {
"type": [
"output_text",
"no_shuffle",
"thinking_tags",
"ans_tags"
],
"ans_candidates": "Move left while yawing right\nMove right while yawing left",
"output_format": "Provide your spatial reasoning inside <thinking></thinking> XML tags, and provide your answer inside <ans></ans> XML tags, e.g... | {
"w/ trap": {
"gt_idx": 0,
"gt_text": "Move right while yawing left",
"prompt": "<input>\nYou are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two f... | {
"scene_name": "scene0791_00",
"seq_name": null
} | [
[
[
-0.94949,
-0.147315,
0.277071,
1.972937
],
[
-0.30901,
0.285268,
-0.907268,
5.137177
],
[
0.054615,
-0.947059,
-0.316381,
2.073987
],
[
0,
0,
0,
1
]
],
[
[
-0.737521,
... | [
[
1165.484009,
0,
654.941589
],
[
0,
1164.543945,
477.277008
],
[
0,
0,
1
]
] | {
"text": [
"pitch up",
"yaw left",
"roll counterclockwise",
"translate right",
"translate up",
"translate forward"
],
"value": [
0.31403504473788685,
-23.332491899923653,
-7.376757460149575,
0.40161718164553983,
-0.1548152188813795,
0.7327035085992879
]
} | 15.200643 | 139.410686 | ||
35 | 15-20 | ScanNet | Move right while yawing left | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | {
"type": [
"output_text",
"no_shuffle",
"thinking_tags",
"ans_tags"
],
"ans_candidates": "Move left while yawing right\nMove right while yawing left",
"output_format": "Provide your spatial reasoning inside <thinking></thinking> XML tags, and provide your answer inside <ans></ans> XML tags, e.g... | {
"w/ trap": {
"gt_idx": 2,
"gt_text": "Move right while yawing left",
"prompt": "<input>\nYou are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two f... | {
"scene_name": "scene0736_00",
"seq_name": null
} | [
[
[
0.446057,
-0.702557,
0.55448,
1.529962
],
[
-0.889386,
-0.278633,
0.362432,
2.78618
],
[
-0.100133,
-0.654812,
-0.74913,
1.57515
],
[
0,
0,
0,
1
]
],
[
[
0.655611,
... | [
[
1163.445068,
0,
653.626038
],
[
0,
1164.793945,
481.600037
],
[
0,
0,
1
]
] | {
"text": [
"pitch up",
"yaw left",
"roll counterclockwise",
"translate right",
"translate up",
"translate forward"
],
"value": [
2.529782498021375,
-6.230702890655921,
-13.621814456215734,
0.2819350769775548,
-0.05920894606363225,
0.033365001170458486
]
} | 19.074109 | 265.943944 | ||
36 | 15-20 | ScanNet | Move left while yawing right | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | {
"type": [
"output_text",
"no_shuffle",
"thinking_tags",
"ans_tags"
],
"ans_candidates": "Move left while yawing right\nMove right while yawing left",
"output_format": "Provide your spatial reasoning inside <thinking></thinking> XML tags, and provide your answer inside <ans></ans> XML tags, e.g... | {
"w/ trap": {
"gt_idx": 1,
"gt_text": "Move left while yawing right",
"prompt": "<input>\nYou are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two f... | {
"scene_name": "scene0804_00",
"seq_name": null
} | [
[
[
0.946883,
0.010614,
-0.321402,
3.145066
],
[
0.321571,
-0.025,
0.946555,
1.657694
],
[
0.002012,
-0.999631,
-0.027085,
1.54369
],
[
0,
0,
0,
1
]
],
[
[
0.912301,
-0... | [
[
1165.484009,
0,
654.941589
],
[
0,
1164.543945,
477.277008
],
[
0,
0,
1
]
] | {
"text": [
"pitch down",
"yaw right",
"roll clockwise",
"translate left",
"translate up",
"translate forward"
],
"value": [
-5.176369135632322,
42.684605125880964,
4.638992807711116,
-0.31097447133801603,
-0.03698607055053649,
0.34243094163453147
]
} | 15.585828 | 456.802264 | ||
37 | 15-20 | ScanNet | Move right while yawing left | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | <input>
You are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two frames. Use the visual difference between them to reason about the camera motion in 3D space.
</in... | {
"type": [
"output_text",
"no_shuffle",
"thinking_tags",
"ans_tags"
],
"ans_candidates": "Move left while yawing right\nMove right while yawing left",
"output_format": "Provide your spatial reasoning inside <thinking></thinking> XML tags, and provide your answer inside <ans></ans> XML tags, e.g... | {
"w/ trap": {
"gt_idx": 0,
"gt_text": "Move right while yawing left",
"prompt": "<input>\nYou are given two consecutive frames from a scene. The first image shows the **source viewpoint**, and the second image shows the **target viewpoint**. The camera moves and rotates simultaneously between these two f... | {
"scene_name": "scene0806_00",
"seq_name": null
} | [
[
[
0.964273,
0.116534,
-0.237902,
3.839192
],
[
0.262243,
-0.547024,
0.794981,
0.903214
],
[
-0.037495,
-0.828966,
-0.55804,
1.534349
],
[
0,
0,
0,
1
]
],
[
[
0.810393,
... | [
[
1165.371094,
0,
650.862061
],
[
0,
1166.109009,
488.595032
],
[
0,
0,
1
]
] | {
"text": [
"pitch up",
"yaw left",
"roll counterclockwise",
"translate right",
"translate up",
"translate forward"
],
"value": [
2.530314475452862,
-16.84361951670938,
-12.041937011910345,
0.6748369511203758,
-0.02494310541293432,
0.007336346569206453
]
} | 15.849558 | 65.444042 |
End of preview. Expand in Data Studio
Split by source dataset:
| 7 Scenes | ScanNet | |
|---|---|---|
| count | 475 | 1038 |
Label distribution (move left.../move right...) splited by angular level (15/30/45/60):
| 15-20 | 30-35 | 45-50 | 60-65 | All | |
|---|---|---|---|---|---|
| Move left while yawing right | 265 | 195 | 167 | 96 | 723 |
| Move right while yawing left | 299 | 218 | 183 | 90 | 790 |
| All | 564 | 413 | 350 | 186 | 1513 |
- Downloads last month
- 73