TopoRetarget: Interaction-Preserving Retargeting for Dexterous Manipulation
2026-06-15 • Robotics
Robotics
AI summaryⓘ
The authors created TopoRetarget, a method that changes human hand movement data so robot hands can use it while keeping important hand-object contacts intact. This is important for teaching robots to do precise hand tasks using reinforcement learning. By focusing on the key points of hands and objects and carefully adjusting them, their approach improves how well robots learn and perform tasks like spinning a pen or moving a cube. Tests show their method works better than others and even works directly on real robot hardware.
dexterous manipulationreinforcement learningretargetinghand-object interactionLaplacian deformationcontact precisionkinematic constraintszero-shot transferrobot handpolicy learning
Authors
Jielin Wu, Shenzhe Yao, Guanqi He, Xiaohan Liu, Zhaoqing Zeng, Xiangrui Jiang, Han Yang, Wentao Zhang, Hang Zhao
Abstract
Human hand-object demonstrations provide dense reference motions for training dexterous manipulation reinforcement learning (RL) policies through reference tracking. However, to use such demonstrations for RL policy learning, retargeting must preserve hand pose and task-relevant hand-object contact structure. Otherwise, contact and feasibility artifacts can degrade downstream RL policy performance. We introduce TopoRetarget, an interaction-preserving retargeting framework that uses a single set of parameters across diverse retargeting conditions while maintaining task-relevant hand-object interaction and adapting human demonstrations to dexterous robot hands. The method constructs a sparse interaction graph over hand and object keypoints and optimizes distance-weighted Laplacian deformation with directional consistency, kinematic constraints, and penetration handling. Evaluations show that the generated references improve both interaction fidelity and policy learning: TopoRetarget achieves the best contact precision and alignment over all baselines on the ContactPose Dataset, improves Pen-Spin training success by 40.6 percentage points over the existing baseline methods, and enables zero-shot transfer to Wuji Hand hardware on cube reorientation and pen spinning.