---
license: mit
pretty_name: DexJoCo preprocessed
tags:
- robotics
- dexterous-manipulation
- point-cloud
- mujoco
---

# DexJoCo, preprocessed

Training frames for [DexJoCo](https://github.com/brave-eai/dexjoco) (Franka + Allegro, MuJoCo, 11 tasks),
re-rendered from the official demos: RGB-D at 224 x 224 with body ids, camera poses, robot state and actions.
MIT, like the DexJoCo data it is made from ([License](#license)).

## Dataset versions

Dataset versions are named **`data-vN`** (tags of this repo; tags never move). Always download a tag and check it with
its manifest.

| dataset | revision | date | demos | GB | change |
|---|---|---|---|---|---|
| `data-v1` | `110a68c5` | 10-06 | 1,059 | 113.5 | first upload (incl. 42 recovered unlock_ipad demos) |
| `data-v2` | `421daed6` | 10-28 | 1,100 | 119.9 | + 41 recovered demos |
| `data-v3` | `72e98046` | 10-28 | 1,100 | 119.7 | 5 frozen-start demos trimmed |
| `data-v4` | tag | 11-01 | 1,100 | 22.3 | `data-v3` in **LeRobot v3** (no H5): RGB H.264 crf 18 (lossy, 31.4 dB worst frame), depth / body id lossless; not yet validated for training |

Manifest sha256 (`find . -name '*.h5' -print0 | LC_ALL=C sort -z | xargs -0 sha256sum | sha256sum`):

| dataset | sha256 |
|---|---|
| `data-v1` | `1dca8061623db35321e1e3f8df73991ed314e398fca9d771f71204cda24ce5a3` |
| `data-v2` | `91c2468b9bd8a411a918914794736c4d294529e3d6d4ea89c92f13422985f7b7` |
| `data-v3` | `6ff159cb75a86051a46a705c13a4d9e22f26a093cdeefbf269c359d524cd9db9` |
| `data-v4` | `3eeda88c370159604998c3ab85dca0ad84bcf191b1a57d528cdea4ca5fc4be0b` (every file in the task folders) |

## Download and check

```bash
hf download chrockey/DexJoCo-preprocessed --repo-type dataset --revision data-v3 --local-dir DexJoCo-preprocessed
hf download chrockey/DexJoCo-preprocessed versions/data-v3.sha256 --repo-type dataset --local-dir manifests
cd DexJoCo-preprocessed && sha256sum -c --quiet ../manifests/versions/data-v3.sha256
```

`--revision data-v4` gives LeRobot v3 (one dataset per task: `<task>/{meta,data,videos,sidecars}`; `lerobot>=0.6.1`
reads RGB + state; depth and body id are lossless FFV1 and need a custom reader); `data-v1`..`data-v3` give H5.
Add `--include "water_plant/*"` for one task. Always pin `--revision`.

## Contents

| task | arms | demos | frames | cameras | GB |
|---|---|---|---|---|---|
| `bimanual_assembly` | 2 | 100 | 72,642 | back, handcam_rgb_left/right | 16.5 |
| `bimanual_hanoi` | 2 | 100 | 109,105 | back, handcam_rgb_left/right | 26.3 |
| `bimanual_microwave_cook` | 2 | 100 | 84,790 | back, handcam_rgb_left/right | 20.6 |
| `bimanual_photograph` | 2 | 100 | 40,255 | back, handcam_rgb_left/right | 8.3 |
| `bimanual_unlock_ipad` | 2 | 100 | 53,144 | back, handcam_rgb_left/right | 12.1 |
| `click_mouse` | 1 | 100 | 39,982 | front, handcam_rgb | 5.4 |
| `fold_glasses` | 1 | 100 | 57,900 | front, handcam_rgb | 8.8 |
| `hammer_nail` | 1 | 100 | 25,034 | front, handcam_rgb | 4.3 |
| `pick_bucket` | 1 | 100 | 46,910 | front, handcam_rgb | 7.3 |
| `pinch_tongs` | 1 | 100 | 43,441 | front, handcam_rgb | 5.9 |
| `water_plant` | 1 | 100 | 27,753 | front, handcam_rgb | 4.3 |

Layout `<task>/<task>_demo_<k>_<date>_<time>.h5` (unlock_ipad files: `ipad_unlock_demo_*`). Counts are for `data-v3`.

## H5 schema (T frames at 30 Hz)

| key | shape | meaning |
|---|---|---|
| attrs | | `task`, `prompt`, `n_arm`, `n_frames`, `size`, `success`, `cameras`; `trim_start` in trimmed demos |
| `actions` | (≤T, 23·n_arm) | per arm `[pos3, quat4, hand16]`, arms interleaved; may stop before T |
| `proprio` | (T, 23·n_arm) | `[tcp pose 7·n_arm \| hand joints 16·n_arm]` (blocked) |
| `qpos` | (T, nq) | MuJoCo `qpos` |
| `body_names`, `is_robot`, `is_hand` | (B,) | bodies and robot / hand masks |
| `xpos`, `xquat` | (T, B, 3/4) | body poses |
| `<cam>/rgb`, `/depth`, `/body_id` | (T, 224, 224, ·) | RGB, depth (m, float16), body id |
| `<cam>/intrinsic`, `/extrinsic` | (3, 3), (T, 4, 4) | camera intrinsics, per-frame pose |

## How it was made

- Simulator `brave-eai/dexjoco` `8d23b0f`; raw demos from `DexJoCo/DexJoCo-Datasets-Raw`.
- Each demo is replayed in the simulator, holding the last action 60 steps (hanoi 90, unlock_ipad 150, photograph 0),
  and rendered at 224 x 224.
- Every demo succeeds on replay. 83 needed recovery: 42 unlock_ipad (wrong iPad pose, brave-eai/dexjoco#27) and 41
  marginal-contact demos (initial object pose search ≤ 3.2 mm).
- `data-v3` trims 5 demos whose recording is frozen at the start (click_mouse 2, microwave_cook 3).
- Checked against the raw recordings.

## License

MIT (`LICENSE`). Made from [`DexJoCo/DexJoCo-Datasets-Raw`](https://huggingface.co/datasets/DexJoCo/DexJoCo-Datasets-Raw)
(MIT, Copyright (c) 2026 Dexjoco contributors) with the [DexJoCo](https://github.com/brave-eai/dexjoco) simulator (MIT).
The rendered robots are the robot models bundled with DexJoCo: Franka Emika Panda (Apache-2.0) and Wonik Allegro hand
(BSD-2-Clause, Copyright (c) 2016, SimLab); no model files are included here. If you use this data, please cite DexJoCo:

```bibtex
@misc{wang2026dexjocobenchmarktoolkittaskoriented,
      title={DexJoCo: A Benchmark and Toolkit for Task-Oriented Dexterous Manipulation on MuJoCo},
      author={Hanwen Wang and Weizhi Zhao and Xiangyu Wang and Siyuan Huang and He Lin and Boyuan Zheng and Rongtao Xu and Gang Wang and Yao Mu and He Wang and Lue Fan and Hongsheng Li and Zhaoxiang Zhang and Tieniu Tan},
      year={2026},
      eprint={2605.16257},
      archivePrefix={arXiv},
      primaryClass={cs.RO}
}
```