Event
OpenJaws: Open-Source UMI Hardware and Stereo Localization Framework
1 reports · 1 sources
Adithyavairavan Murali and co-authors released OpenJaws, a fully open-source sensorized UMI software and hardware framework that uses a modular 3D-printed gripper paired with a low-cost stereo camera to recover metric camera trajectories, improving the localization accuracy of action labels in handheld demonstrations. The authors also propose UMI-Loc-Bench-v1, a benchmark for quantifying the real-world localization accuracy of handheld data-collection devices, and state that stereo localization is clearly superior to monocular approaches. The authors demonstrate that policies trained on data collected with OpenJaws have been deployed on real robots across multiple tasks. The hardware, software, and data are released under a commercially friendly license.
Report timeline
- OpenJaws: An open-source hardware UMI framework for generalist manipulation, using stereo localization to improve demonstration accuracy
OpenJaws is a fully open-source sensorized UMI hardware and software framework that pairs a modular 3D-printed gripper with a low-cost stereo camera to recover metric camera trajectories, improving action-label localization accuracy in handheld demonstrations. The authors also propose UMI-Loc-Bench-v1, a benchmark for quantifying the real-world localization accuracy of handheld data-collection devices, and show that stereo localization clearly outperforms monocular alternatives. Policies trained on data collected with OpenJaws have been deployed on real robots across multiple tasks, and the hardware, software, and data are released under a commercially friendly license.
arXiv cs.RO ·