Official implementation of [RSS 2025] Dexonomy: Synthesizing All Dexterous Grasp Types in a Grasp Taxonomy.
Project Page | Paper | Dataset | Annotation UI Demo | Learning
Our algorithm synthesizes contact-rich, penetration-free, and physically plausible dexterous grasps for:
- Any grasp type
- Any object
- Any articulated hand
All starting from just one human-annotated template per hand and grasp type.
The full code of our proposed synthesis pipeline will be released late this year.
This work and the dataset are licensed under CC BY-NC 4.0.
If you find this work useful for your research, please consider citing:
@article{chen2025dexonomy,
title={Dexonomy: Synthesizing All Dexterous Grasp Types in a Grasp Taxonomy},
author={Chen, Jiayi and Ke, Yubin and Peng, Lin and Wang, He},
journal={Robotics: Science and Systems},
year={2025}
}