Search papers, labs, and topics across Lattice.
This paper introduces AnyDexRT, a calibration-free method for retargeting human hand motions to dexterous robotic hands, addressing the limitations of existing techniques that rely on hand-specific tuning and calibration. By leveraging self-supervised fingertip correspondence learning and few-shot human guidance, the method anchors the mapping in task-relevant regions and refines pinch-related poses through a contact classifier. Experimental results demonstrate that AnyDexRT significantly enhances retargeting quality and control intuitiveness across various dexterous hands and real-world tasks, outperforming previous approaches.
Calibration-free dexterous hand retargeting achieves intuitive control and superior performance without the need for hand-specific tuning.
Teleoperation is a key interface for controlling dexterous robotic hands and collecting demonstrations for imitation learning. Its effectiveness largely depends on kinematic retargeting, which maps operator hand motions to feasible and intuitive robot hand motions. Existing methods often require hand-crafted objectives, precise calibration, or global shape matching between human and robot hand spaces, making them sensitive to hand-specific tuning and less reliable across different dexterous hands. We propose AnyDexRT, a calibration-free retargeting method for intuitive dexterous teleoperation across human-like dexterous hands. AnyDexRT combines self-supervised fingertip correspondence learning with few-shot human guidance to anchor the mapping in task-relevant regions, and further refines pinch-related poses using a contact classifier. Experiments on diverse dexterous hands and real-world teleoperation tasks show that AnyDexRT improves retargeting quality, reduces manual tuning, and provides more intuitive and efficient control than prior retargeting methods. Project website: https://chenxi-wang.github.io/projects/anydexrt
