I am working on embodied intelligence. My work on focused on multimodality. Real-time dynamic / static scene understanding, and motion control from language commands.
- ๐ญ Iโm currently working on scene topology understanding.
- ๐ฑ Iโm currently learning large perception multi-modal geometric pretraining.
- ๐ฏ Iโm looking to collaborate on anything towards AGI, especially for geometric understanding and end-to-end perception-planning-control.