Running
7
EgoInfinity Browser
🪞
Static viewer for the EgoInfinity dataset
AI & ML interests
Robotics, especially manipulation and physical interactions
Recent Activity
View all activity
Papers
RoboTok: An Internet-Scale Data Engine for Human Demonstration Retrieval and Dexterous Manipulation Learning
Zero-Shot Sim-to-Real Robot Learning: A Dexterous Manipulation Study on Reactive Catching
Organization Card
Rice RobotΠ Lab
Where robots Physically Interact with the world.
The Rice RobotΠ Lab at Rice University, led by Prof. Kaiyu Hang, studies how robots can physically interact with an uncertain, contact-rich world. Our work spans manipulation planning, interactive perception, learning from in-the-wild video, and benchmarking infrastructure for the field.
What you'll find here
This Hugging Face organization hosts the lab's open-source datasets, interactive demos, and model artifacts. For papers, members, and full project listings, please visit our lab website.
On Hugging Face
- EgoInfinity (coming soon): Large-scale automated reconstruction of 3D hand-object interactions from in-the-wild egocentric video, designed for scaling up robot training data.
Other ongoing lab projects
- ManiDreams: Open-source modular framework for uncertainty-aware manipulation planning over intuitive physics models.
- ManipulationNet: Community-driven global infrastructure for benchmarking robot manipulation research at scale.
Links
- 🌐 Lab website
- 💻 GitHub
- 📝 Publications
- 📨 Join the lab