DexMimicGen
NVIDIA system that synthesizes ~21K bimanual dexterous sim demos from 60 human demos across 9 tasks; real GR-1 transfer.
| Year | 2024 |
|---|---|
| Episodes | 21,000 |
| Embodiments | franka_panda, humanoid_other, parallel_jaw |
| Modalities | depth, ee_pose, proprioception, rgb |
| Task categories | manipulation |
| License | CC-BY-NC-4.0 |
| Access | open |
| Maintainer | NVIDIA Research (with UT Austin and UC San Diego) |
DexMimicGen is an automated data-generation system that synthesizes large-scale trajectory datasets for bimanual dexterous manipulation from a handful of human teleoperation demonstrations. It generated over 21,000 demonstrations from 60 source human demos across 9 simulation tasks in robosuite/MuJoCo, spanning three embodiments (bimanual Panda arms with parallel-jaw grippers, bimanual Panda arms with dexterous hands, and a Fourier GR-1 humanoid with Inspire dexterous hands), and includes a Real2Sim2Real pipeline validated on a real GR-1 robot.