DEXTEROUS MANIPULATION

The world of dexterous hands.

Compare hardware. Discover applications. Share what works. An open reference and community for dexterous manipulation.

ObjDex

cypypccpy

Object-centric dexterous manipulation policies learned from human interaction motion.

What you can explore

Explore ARCTIC-based training examples for interacting with articulated objects using a simulated hand.

Before you use it

The release uses legacy Isaac Gym and Python 3.8. Source ARCTIC data and object models retain separate access and use restrictions.

Original contributors

Yuanpei Chen, Chen Wang, Yaodong Yang and C. Karen Liu

License / access: MIT stated in README · ARCTIC and external assets have separate terms

Original code & documentation

Research paper

Keep exploring

ARCTIC

Two human hands interacting with articulated objects, with motion and contact information for reconstruction research.

DexMachina

Physics-based retargeting of human demonstrations into coordinated robot-hand manipulation of articulated objects.

DexTrack

Dexterous manipulation through reinforcement learning that tracks human-derived hand–object motion.

Editorial reference · checked 2026-09-22

Start a discussion

Published posts are public. First contributions are reviewed. Only share material you have permission to disclose.

Report a post