Selected work
Research
Statistical ideas for learning systems that are more understandable, reliable, and scalable.
IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · 2025
LATTE-MV: Learning to Anticipate Table Tennis Hits from Monocular Videos
A scalable vision system for reconstructing table-tennis matches from monocular video and training an uncertainty-aware controller to anticipate an opponent's actions.
Project pageYou can also find my work on Google Scholar ↗.