Open-H-Embodiment is a community-driven, multi-embodiment dataset for AI autonomy models in surgical robotics and ultrasound. It contains paired kinematics and video data across tasks like tabletop exercises and clinical procedures, including simulations. The dataset is hosted by NVIDIA on Hugging Face and was last updated on April 3, 2026.
Use Cases
- Train AI models for surgical robotics based on paired kinematics and video data.
- Evaluate robotic autonomy in clinical procedures using the multi-embodiment collection.
- Develop models for ultrasound robotics based on the described task data.
- Simulate and test healthcare robotics applications using the included simulation data.
Strengths
- Dataset is community-driven, suggesting diverse contributions.
- Focus on healthcare robotics applications like surgery and ultrasound.
- Includes both real-world clinical procedures and simulations.
- Last updated on April 3, 2026, indicating recent maintenance.
Limitations
- Column-level documentation is absent; field semantics must be inferred after download.
- Row count, file formats, and license are unknown, which may limit suitability assessment.
- Data may reflect geographic or procedural bias inherent to the source platforms.
Provenance
- Source
- NVIDIA, hosted on Hugging Face.
- Collection Method
- Community-driven collection of LeRobot datasets.
- Time Range
- null
- Freshness
- Last updated 2026-04-03 15:25:37.
- Geography
- null