DexCanvas: Bridging Human Demonstrations and Robot Learning for Dexterous Manipulation
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866917035360911360 |
|---|---|
| author | Xu, Xinyue Sun, Jieqiang Jing Dai Chen, Siyuan Ma, Lanjie Sun, Ke Zhao, Bin Yuan, Jianbo Yi, Sheng Zhu, Haohua Lu, Yiwen |
| author_facet | Xu, Xinyue Sun, Jieqiang Jing Dai Chen, Siyuan Ma, Lanjie Sun, Ke Zhao, Bin Yuan, Jianbo Yi, Sheng Zhu, Haohua Lu, Yiwen |
| contents | We present DexCanvas, a large-scale hybrid real-synthetic human manipulation dataset containing 7,000 hours of dexterous hand-object interactions seeded from 70 hours of real human demonstrations, organized across 21 fundamental manipulation types based on the Cutkosky taxonomy. Each entry combines synchronized multi-view RGB-D, high-precision mocap with MANO hand parameters, and per-frame contact points with physically consistent force profiles. Our real-to-sim pipeline uses reinforcement learning to train policies that control an actuated MANO hand in physics simulation, reproducing human demonstrations while discovering the underlying contact forces that generate the observed object motion. DexCanvas is the first manipulation dataset to combine large-scale real demonstrations, systematic skill coverage based on established taxonomies, and physics-validated contact annotations. The dataset can facilitate research in robotic manipulation learning, contact-rich control, and skill transfer across different hand morphologies. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_15786 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | DexCanvas: Bridging Human Demonstrations and Robot Learning for Dexterous Manipulation Xu, Xinyue Sun, Jieqiang Jing Dai Chen, Siyuan Ma, Lanjie Sun, Ke Zhao, Bin Yuan, Jianbo Yi, Sheng Zhu, Haohua Lu, Yiwen Robotics Machine Learning We present DexCanvas, a large-scale hybrid real-synthetic human manipulation dataset containing 7,000 hours of dexterous hand-object interactions seeded from 70 hours of real human demonstrations, organized across 21 fundamental manipulation types based on the Cutkosky taxonomy. Each entry combines synchronized multi-view RGB-D, high-precision mocap with MANO hand parameters, and per-frame contact points with physically consistent force profiles. Our real-to-sim pipeline uses reinforcement learning to train policies that control an actuated MANO hand in physics simulation, reproducing human demonstrations while discovering the underlying contact forces that generate the observed object motion. DexCanvas is the first manipulation dataset to combine large-scale real demonstrations, systematic skill coverage based on established taxonomies, and physics-validated contact annotations. The dataset can facilitate research in robotic manipulation learning, contact-rich control, and skill transfer across different hand morphologies. |
| title | DexCanvas: Bridging Human Demonstrations and Robot Learning for Dexterous Manipulation |
| topic | Robotics Machine Learning |
| url | https://arxiv.org/abs/2510.15786 |