Dataset Viewer
Auto-converted to Parquet Duplicate
episode_index
int64
0
299
key
stringclasses
4 values
rgb_mae
float64
1.96
3.56
bgr_mae
float64
10.7
47.8
plain_rgb_mae
float64
10.1
11.8
⌀
0
observation.images.head
2.739871
13.147957
null
0
observation.images.wrist
3.105074
15.434131
null
0
observation.images.tactile_left
2.391658
26.894201
10.709073
0
observation.images.tactile_right
2.407054
27.135047
10.765025
1
observation.images.head
2.622159
13.09353
null
1
observation.images.wrist
2.77632
15.687515
null
1
observation.images.tactile_left
2.222776
26.476518
10.446905
1
observation.images.tactile_right
2.236706
26.967782
10.33059
2
observation.images.head
2.7512
13.114899
null
2
observation.images.wrist
3.093771
15.749452
null
2
observation.images.tactile_left
2.366228
26.795705
10.463092
2
observation.images.tactile_right
2.369986
27.266389
10.4883
3
observation.images.head
2.595156
13.126991
null
3
observation.images.wrist
2.901881
15.788335
null
3
observation.images.tactile_left
2.327046
26.727051
10.469651
3
observation.images.tactile_right
2.314187
27.212072
10.41184
4
observation.images.head
2.740123
13.349395
null
4
observation.images.wrist
3.057605
15.86423
null
4
observation.images.tactile_left
2.366535
26.746408
10.533302
4
observation.images.tactile_right
2.364204
27.203978
10.492511
5
observation.images.head
2.742156
13.1129
null
5
observation.images.wrist
3.112433
15.644213
null
5
observation.images.tactile_left
2.371727
26.7668
10.540815
5
observation.images.tactile_right
2.386943
27.229255
10.530812
6
observation.images.head
2.690628
13.167218
null
6
observation.images.wrist
3.202301
15.563201
null
6
observation.images.tactile_left
2.445909
26.814303
10.172777
6
observation.images.tactile_right
2.436007
27.467763
10.12582
7
observation.images.head
2.570369
13.061798
null
7
observation.images.wrist
2.925427
15.612205
null
7
observation.images.tactile_left
2.310346
26.552065
10.324348
7
observation.images.tactile_right
2.339952
27.136845
10.288057
8
observation.images.head
2.718591
13.111731
null
8
observation.images.wrist
3.099615
15.730411
null
8
observation.images.tactile_left
2.374418
26.794873
10.318471
8
observation.images.tactile_right
2.346383
27.442789
10.24412
9
observation.images.head
2.737538
13.267354
null
9
observation.images.wrist
3.233164
15.480227
null
9
observation.images.tactile_left
2.470773
26.915018
10.202047
9
observation.images.tactile_right
2.474243
27.507027
10.204122
10
observation.images.head
2.624348
13.134249
null
10
observation.images.wrist
2.77212
15.074634
null
10
observation.images.tactile_left
2.222112
26.906204
10.533367
10
observation.images.tactile_right
2.228673
27.200591
10.60349
11
observation.images.head
2.739033
13.095778
null
11
observation.images.wrist
3.240737
15.76969
null
11
observation.images.tactile_left
2.477358
26.95491
10.63524
11
observation.images.tactile_right
2.500719
27.195199
10.670421
12
observation.images.head
2.59366
13.04662
null
12
observation.images.wrist
2.884775
15.493064
null
12
observation.images.tactile_left
2.318676
26.698764
10.187869
12
observation.images.tactile_right
2.30769
27.341084
10.168012
13
observation.images.head
2.740416
13.219023
null
13
observation.images.wrist
3.113594
15.661428
null
13
observation.images.tactile_left
2.371021
26.986131
10.338102
13
observation.images.tactile_right
2.363442
27.586685
10.239223
14
observation.images.head
2.605822
13.087259
null
14
observation.images.wrist
2.56823
15.136841
null
14
observation.images.tactile_left
2.391471
26.940603
10.615624
14
observation.images.tactile_right
2.416249
27.155424
10.706403
15
observation.images.head
2.38863
13.091845
null
15
observation.images.wrist
2.276398
15.565894
null
15
observation.images.tactile_left
2.377255
26.811364
10.582954
15
observation.images.tactile_right
2.383508
27.080658
10.615302
16
observation.images.head
2.582412
13.112242
null
16
observation.images.wrist
2.899667
15.263618
null
16
observation.images.tactile_left
2.456921
26.988296
10.322607
16
observation.images.tactile_right
2.430571
27.585227
10.202199
17
observation.images.head
2.749416
13.155136
null
17
observation.images.wrist
3.101105
15.401227
null
17
observation.images.tactile_left
2.369126
26.916241
10.338538
17
observation.images.tactile_right
2.346568
27.588368
10.22709
18
observation.images.head
2.747233
13.13131
null
18
observation.images.wrist
3.081098
15.708192
null
18
observation.images.tactile_left
2.41488
26.957738
10.723273
18
observation.images.tactile_right
2.40548
27.143871
10.767772
19
observation.images.head
2.737363
13.128265
null
19
observation.images.wrist
3.090255
15.676874
null
19
observation.images.tactile_left
2.383048
26.850485
10.449039
19
observation.images.tactile_right
2.405506
27.345437
10.467612
20
observation.images.head
2.715141
13.125002
null
20
observation.images.wrist
3.072632
15.614703
null
20
observation.images.tactile_left
2.352337
26.58532
10.368388
20
observation.images.tactile_right
2.350012
27.0927
10.308243
21
observation.images.head
2.725501
13.143383
null
21
observation.images.wrist
3.233811
15.675747
null
21
observation.images.tactile_left
2.525174
26.92774
10.737004
21
observation.images.tactile_right
2.521337
27.145946
10.771953
22
observation.images.head
2.724864
13.496543
null
22
observation.images.wrist
3.279887
16.412094
null
22
observation.images.tactile_left
2.486479
26.840518
10.506826
22
observation.images.tactile_right
2.507922
27.242662
10.549342
23
observation.images.head
2.730639
13.108568
null
23
observation.images.wrist
3.086764
15.512071
null
23
observation.images.tactile_left
2.353135
26.816399
10.243002
23
observation.images.tactile_right
2.331651
27.465849
10.211463
24
observation.images.head
2.623704
13.098544
null
24
observation.images.wrist
2.783765
15.666434
null
24
observation.images.tactile_left
2.226791
26.777142
10.744499
24
observation.images.tactile_right
2.227633
26.9946
10.750615
End of preview. Expand in Data Studio

OpenViTac Benchmark Dataset

OpenViTac is a visuo-tactile robot manipulation benchmark covering physical-property perception, fragile-object handling, contact-rich manipulation, and precise insertion. This release contains 1,500 simulated episodes in LeRobot v2.1 format across 11 tasks: 1,300 with GelSight mini and 100 each with Neote and Xense for insert_USB.

For the benchmark and data-collection details, see the paper and code repository.

Simulation tasks

Each Simulation/<sensor>/<task>/ directory is a separate LeRobot dataset. Its meta/ directory contains episode metadata, data/chunk-000/ contains one Parquet file per episode, and videos/chunk-000/ contains the corresponding camera and tactile videos. GelSight mini data is under Simulation/GelSight_mini/.

Capability Task directory Task summary Episodes
Physical-property perception weight_classify Identify whether a block is light or heavy through interaction and place it on the matching target. 100
Physical-property perception hardness_classify Identify whether a block is soft or hard through contact and place it on the matching target. 100
Physical-property perception roughness_classify Identify whether a block is smooth or rough through contact and place it on the matching target. 100
Physical-property perception roughness_guided_regrasp Use surface roughness to decide whether to keep the grasped block or regrasp another one. 100
Physical-property perception empty_can_selection Distinguish an empty can from visually similar cans through interaction and place it in a basket. 100
Fragility-aware manipulation grasp_chip Grasp and move a fragile chip without damaging it. 100
Contact-rich manipulation gear_assembly Mount and rotate a gear while maintaining engagement with a paired gear. 100
Contact-rich manipulation pull_drawer Maintain a stable grasp on the handle while opening a drawer. 100
Precision manipulation insert_USB Align a USB connector with its slot and insert it. 100
Precision manipulation insert_block_v1 Align and insert a designated block into the matching opening. 100
Precision manipulation insert_block_v2 Select the instructed object, then insert it into the matching opening. 300

The insert_USB task also includes Neote and Xense tactile sensor data, with 100 episodes each in Simulation/Neote/insert_USB/ and Simulation/Xense/insert_USB/.

Real-world tasks

Downloads last month
665

Paper for FVL-Lab/OpenViTac