Dataset Viewer
Auto-converted to Parquet Duplicate
Search is not available for this dataset
video
video
11.7
36.1
label
class label
99 classes
0000_75c22d4d1d52e994
0000_75c22d4d1d52e994
0000_75c22d4d1d52e994
0000_75c22d4d1d52e994
0000_75c22d4d1d52e994
1001_0974e10be979fd68
1001_0974e10be979fd68
1001_0974e10be979fd68
1001_0974e10be979fd68
1001_0974e10be979fd68
2002_70f7fffaedcb1975
2002_70f7fffaedcb1975
2002_70f7fffaedcb1975
2002_70f7fffaedcb1975
2002_70f7fffaedcb1975
3003_67b13ee4fd177c9d
3003_67b13ee4fd177c9d
3003_67b13ee4fd177c9d
3003_67b13ee4fd177c9d
3003_67b13ee4fd177c9d
4004_b3569c3ba462f27b
4004_b3569c3ba462f27b
4004_b3569c3ba462f27b
4004_b3569c3ba462f27b
4004_b3569c3ba462f27b
5005_24d13ee7e5954e80
5005_24d13ee7e5954e80
5005_24d13ee7e5954e80
5005_24d13ee7e5954e80
5005_24d13ee7e5954e80
6006_be2ebd9b8e81fee2
6006_be2ebd9b8e81fee2
6006_be2ebd9b8e81fee2
6006_be2ebd9b8e81fee2
6006_be2ebd9b8e81fee2
7007_7db049140e504397
7007_7db049140e504397
7007_7db049140e504397
7007_7db049140e504397
7007_7db049140e504397
8008_f4e03d1d38a38d91
8008_f4e03d1d38a38d91
8008_f4e03d1d38a38d91
8008_f4e03d1d38a38d91
8008_f4e03d1d38a38d91
9009_a8914760f2731175
9009_a8914760f2731175
9009_a8914760f2731175
9009_a8914760f2731175
9009_a8914760f2731175
10010_a32ec459f2e900f4
10010_a32ec459f2e900f4
10010_a32ec459f2e900f4
10010_a32ec459f2e900f4
10010_a32ec459f2e900f4
11011_39092d8f94386ef5
11011_39092d8f94386ef5
11011_39092d8f94386ef5
11011_39092d8f94386ef5
11011_39092d8f94386ef5
12012_6a27bf05e4118feb
12012_6a27bf05e4118feb
12012_6a27bf05e4118feb
12012_6a27bf05e4118feb
12012_6a27bf05e4118feb
13013_d938a13c20634c86
13013_d938a13c20634c86
13013_d938a13c20634c86
13013_d938a13c20634c86
13013_d938a13c20634c86
14014_a528af3ae6bd8875
14014_a528af3ae6bd8875
14014_a528af3ae6bd8875
14014_a528af3ae6bd8875
14014_a528af3ae6bd8875
15015_658dc401e0c16a93
15015_658dc401e0c16a93
15015_658dc401e0c16a93
15015_658dc401e0c16a93
15015_658dc401e0c16a93
16016_3c58d56017646fd8
16016_3c58d56017646fd8
16016_3c58d56017646fd8
16016_3c58d56017646fd8
16016_3c58d56017646fd8
17017_01a90396b16ed52c
17017_01a90396b16ed52c
17017_01a90396b16ed52c
17017_01a90396b16ed52c
17017_01a90396b16ed52c
18018_93471f038e9e6503
18018_93471f038e9e6503
18018_93471f038e9e6503
18018_93471f038e9e6503
18018_93471f038e9e6503
19019_93ecfec4082a05ef
19019_93ecfec4082a05ef
19019_93ecfec4082a05ef
19019_93ecfec4082a05ef
19019_93ecfec4082a05ef
End of preview. Expand in Data Studio

V10.11c mixed mesh/primitive clutter — 99 episodes

Clean pick-and-place demonstrations from pact_place_corridor_v10_11c_33pct_taller_primitives, each carrying a fixed exterior table camera (exo_camera_1) with per-frame calibration.

What the scene contains

Six table-clutter bodies: three mesh assets and three runtime MuJoCo primitives. The primitives are built on the episode's MjSpec rather than compiled into the scene, so the certified scene files stay byte-identical.

  • 01 tall_route_cylinder — primitive
  • 03 plate_10 — mesh
  • 04 plate_22 — mesh
  • 06 soap_bottle_11 — mesh
  • 08 tall_near_target_cylinder — primitive
  • 09 tall_near_target_box — primitive

Slots 08/09 are sampled per episode in a bounded annular sector around the target cup, so near-target clutter varies from episode to episode. The remaining clutter uses the frozen V9.5 layout with its inherited vessel jitter.

Every row is a clean success

Each published episode is a strict-clean success under the collection's own predicate: the task succeeded and there was no contact with clutter, the hazard bar, mounted fixtures or the rest of the environment. Rejected attempts are not published as rows, but they remain in ledger.jsonl so the yield is auditable (99 accepted from 481 attempts).

The exterior camera

Every row carries:

  • episode_00000000_exo_camera_1.mp4 — exterior RGB, plus a depth track;
  • traj_0/obs/sensor_param/exo_camera_1 in trajectory.h5, holding extrinsic_cv (T,3,4), cam2world_gl (T,4,4) and intrinsic_cv (T,3,3) as finite float64;
  • a fixed pose within each episode — the within-episode max delta of all three calibration arrays is exactly 0.0.

The pose is re-drawn between episodes, so load the extrinsic from the episode you are projecting. Reusing one episode's extrinsic_cv across the set will silently misproject the others, since the intrinsics do match.

manifest.json carries that validation per row plus a SHA-256 for every file.

Layout

pact_place_corridor_v10_11c_100/
  manifest.json        per-row metadata, camera validation and file hashes
  ledger.jsonl         the full run ledger, including rejected attempts
  contract.json        the frozen run contract
  closeout.json        the run close-out
  rows/NNN_<row>/      one directory per accepted episode

Per-row files: trajectory.h5, result.json, and MP4s for the exterior camera (RGB + depth), the wrist camera (RGB + depth) and the proximity-sensor heatmap.

Coverage

24 cells: four layout families x two intrusion sides x three pendant poses. Cells, seeds and per-episode contact totals are in manifest.json.

Downloads last month
34