DexSeed: Expanding Human Hand-Object Interactions into Robot Demonstrations for Dexterous Policy Learning
DexSeed: Expanding Human Hand-Object Interactions into Robot Demonstrations for Dexterous Policy Learning
This research presents DexSeed, a framework for low-cost generation of dexterous robot data from human hand-object interactions. The method organizes source sequences by grasp mode, uses InteractionMesh-based retargeting and arm trajectory augmentation to adapt sequences to robot hands, object geometries, and locations while preserving local relationships. An RL finger-control policy produces executable rollouts, and rendered observations paired with grasp descriptions fine-tune a VLA. Experiments show improved contact fidelity and reliable whole-arm execution in simulation and real-world deployment, enabling the execution of language-specified grasps using only generated demonstrations.
Source: arXiv Robotics — research abstracts · Read original article ↗
Loading article text…
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
- Environment
- SimulationOpen source S4
- code: not_reported
- weights: not_reported
- dataset: not_reported
- paper: not_reported
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
ns under arm-tracking dynamics in physics simulation. Specifically, we organize source sequences by grasp mode with shared language descriptions. InteractionMesh-based retargeting and arm trajectory augmentation adapt these sequences to the robot hand, different object geometries, and pick-and-place locations while preserving local hand--object relationships. An RL finger-control policy produces executable rollouts,
Open source S4
ions can execute language-specified grasps. Project page: https://robbyant-research.github.io/DexSeed/.
Open source S6
Source：arXiv Robotics — research abstracts · arxiv.org