SynManDex: Synthesizing Human-like Dexterous Grasps from Synthetic Human Pre-Grasps
2026-06-08 • Robotics
Robotics
AI summaryⓘ
The authors created SynManDex, a system that helps robots learn how to grasp and use objects by starting from human hand movements and adjusting them to fit robotic hands. It first generates human-like hand poses near objects, then refines these poses to make sure the robot can hold the objects securely and naturally. Their system works well in both simulations and real robots, successfully handling tasks like pouring tea and playing the flute. This approach balances strong grasping ability with movements that look human-like.
human hand-object interactionrobotic graspingdexterous manipulationforce-closurepose retargetingsynthetic data generationbimanual roboticsgrasp stabilityvisual language models (VLM)trajectory optimization
Authors
Yanming Shao, Zanxin Chen, Wenwei Lin, Mingjie Zhou, Tianxing Chen, Xiaokang Yang, Yichen Chi, Yao Mu
Abstract
Human hand-object interactions encode functional intent, but direct transfer to robotic hands often fails under morphology, contact, and reachability constraints. We present SynManDex, a synthetic pipeline that uses generated human pre-grasps as affordance-aware proposals and resolves the final contacts with robot-native optimization. SynManDex samples object-conditioned digital human pre-grasps, retargets them to dexterous robotic hand poses, optimizes force-closure contacts on the target embodiment, and admits trajectories that pass checks from each step. The resulting keyframes support both grasp-and-lift demonstrations and various prehensile manipulation tasks such as tea pouring, photo taking, and flute playing, designed via VLM agents. As a result, SynManDex combines high grasp quality (86.4\% grasp stability) with 4.67/5 human-likeness (93.4\%). It achieves 80.7\% successes in simulation and 25/30 (83.3\%) real-robot successes when applied to a 36-DOF bimanual dexterous robotic platform.