Reproduce “DexTwist: Dexterous Hand Retargeting for Twist Motion via Mixed Reality-based Teleoperation” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Dongmyoung Lee, Chengxi Li, Dongheui Lee
DexTwist addresses remote hand control by preserving object turning when human and robot hands differ. It detects a tripod pinch, tracks accumulated turning progress, and refines robot finger motion while stabilizing contacts. In real-world trials, its object-rotation RMSE was 14.3°, versus 23.2° for Vector Retargeting. Simulation screw-axis deviation fell from 13.6° to 6.1°; real-world axis stability was not measured.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.