{"success":true,"data":{"author":"Chen Wang et al. (Stanford University)","bucketAccessType":"public","bucketName":"chenwangj/DexCap-Data","bucketPrefix":"","category":"Robotics","cloudProvider":"huggingface","createdAt":"2026-11-14T14:15:10.614Z","description":"The official dataset collected with DexCap, a portable hand motion capture system for learning dexterous robot manipulation from human demonstrations (Wang et al., arXiv:2403.07788). It contains 30 minutes of mocap data for a wiping task and 60 minutes of in-the-wild mocap data for a packaging task. Raw data are zip archives in which each frame folder holds the chest camera RGB image (JPEG) and depth image (PNG), the chest camera 6-DoF pose in the world frame, left and right hand 6-DoF poses, and 3D hand joint positions in the palm frame as text files. Postprocessed training sets are HDF5 files with 10,000-point point clouds and a 5-step action gap, readable with h5py and the training scripts in the DexCap GitHub repository. Measured 2026-11-14 at about 108 GB across two raw zips and two HDF5 files. Useful for human-to-robot retargeting, point-cloud imitation learning and dexterous bimanual manipulation. The card releases the dataset under CC BY 4.0; cite the DexCap paper.","displayTitle":"DexCap: Hand Motion Capture Data for Dexterous Manipulation","id":"69246c17-6a91-58d3-bfea-fd836ab2a562","license":"CC-BY-4.0","licenseEvidenceUrl":"https://huggingface.co/datasets/chenwangj/DexCap-Data","objectCountEstimate":6,"probeStatus":"ok","referenceUrl":"https://arxiv.org/abs/2403.07788","region":"us","schemaCard":{"accessHints":[{"language":"python-datasets","snippet":"from datasets import load_dataset\nds = load_dataset(\"chenwangj/DexCap-Data\")"},{"language":"python-hub","snippet":"from huggingface_hub import snapshot_download\nsnapshot_download(repo_id=\"chenwangj/DexCap-Data\", repo_type=\"dataset\")"}],"caveats":["HuggingFace metadata: author=chenwangj, license:cc-by-4.0, downloads=768"],"format":"mixed","generatedAt":"2026-11-14T14:15:18.023Z","layout":{"example":"hand_packaging_wild_1-20_5actiongap_10000points.hdf5","pattern":"{repoId}/{filename}"},"mountPath":"/mnt/archive","sampleHeaders":[{"bytes":"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","encoding":"hex","key":"hand_packaging_wild_1-20_5actiongap_10000points.hdf5","truncatedAt":512}],"sampleObjects":["hand_packaging_wild_1-20_5actiongap_10000points.hdf5","hand_wiping_1-14_5actiongap_10000points.hdf5","rawdata_packaging_wild_1-20.zip","rawdata_wipe_1-14.zip"],"sidecars":[{"bytes":"---\nlicense: cc-by-4.0\n---\n# Dataset Card for DexCap-Data\n\n## Table of Contents\n- [Table of Contents](#table-of-contents)\n- [Dataset Description](#dataset-description)\n  - [Dataset Summary](#dataset-summary)\n- [Dataset Structure](#dataset-structure)\n- [Dataset Creation](#dataset-creation)\n- [Additional Information](#additional-information)\n  - [Licensing Information](#licensing-information)\n  - [Citation Information](#citation-information)\n\n## Dataset Description\n\n- **Homepage:** https://dex-cap.github.io/\n- **Repository:** https://github.com/j96w/DexCap\n- **Paper:** https://arxiv.org/abs/2403.07788\n\n### Dataset Summary\n\nThis is the official dataset collected by the DexCap system to train dexterous robot manipulation using human hand motion capture data, as presented in the [paper](https://arxiv.org/abs/2403.07788). It contains 30 minutes of mocap data for the wiping task and 60 minutes of in-the-wild mocap data for the packaging task.\n\n## Dataset Structure\n\nBoth raw data (`.zip`) and postprocessed data (`.hdf5`) are provided. The raw data is structured as follows:\n```\nsave_data_scenario_1\n├── frame_0\n│   ├── color_image.jpg           # Chest camera RGB image\n│   ├── depth_image.png           # Chest camera depth image\n│   ├── pose.txt                  # Chest camera 6-DoF pose in world frame\n│   ├── pose_2.txt                # Left hand 6-DoF pose in world frame\n│   ├── pose_3.txt                # Right hand 6_DoF pose in world frame\n│   ├── left_hand_joint.txt       # Left hand joint positions (3D) in the palm frame\n│   └── right_hand_joint.txt      # Right hand joint positions (3D) in the palm frame\n├── frame_1\n└── ...\n```\nDetails on how to postprocess the raw data into a training dataset (`.hdf5`) are introduced in the [GitHub repo](https://github.com/j96w/DexCap). The training scripts for policy learning with the HDF5 dataset are also included in the repo.\n\n## Dataset Creation\n\nAll data are collected by the [DexCap system](https://dex-cap.github.io/).\n\n## Additional Information\n\n### Licensing Information\n\nThis dataset is released under the [Creative Commons Attribution 4.0 International (CC BY 4.0)](https://creativecommons.org/licenses/by/4.0/legalcode) license.\n\n### Citation Information\n\n```bibtex\n@article{wang2024dexcap,\n  title = {DexCap: Scalable and Portable Mocap Data Collection System for Dexterous Manipulation},\n  author = {Wang, Chen and Shi, Haochen and Wang, Weizhuo and Zhang, Ruohan and Fei-Fei, Li and Liu, C. Karen},\n  journal = {arXiv preprint arXiv:2403.07788},\n  year = {2024}\n}\n```","encoding":"utf8","key":"README.md","name":"README.md","size":2621,"truncated":false}],"sourceObjectKey":"hand_packaging_wild_1-20_5actiongap_10000points.hdf5","starterCell":"from huggingface_hub import hf_hub_download\npath = hf_hub_download(repo_id=\"chenwangj/DexCap-Data\", filename=\"hand_packaging_wild_1-20_5actiongap_10000points.hdf5\", repo_type=\"dataset\")\nprint(path)","starterSamplePath":"hand_packaging_wild_1-20_5actiongap_10000points.hdf5","totalObjects":6},"summary":"The official dataset collected with DexCap, a portable hand motion capture system for learning dexterous robot manipulation from human demonstrations (Wang et al.,…","title":"DexCap: Hand Motion Capture Data for Dexterous Manipulation","updatedAt":"2026-11-14T17:40:05.815Z","visibility":"public","websiteUrl":"https://dex-cap.github.io/"}}