I am a postdoctoral researcher at the GRASP Lab under the supervision of Prof. Kostas Daniilidis working on the intersection of computer vision, robotics and machine learning with and without event cameras.

My research focuses on event-based sensing, and recently extended from vision-based (i.e., event cameras) to IMUs as part of my recent work on Lie Events for IMU-based navigation, and event-based computation in the embedding space via asynchronous neural networks. I plan to extend this framework to Predictive Event Tokenization (PET), generalizing event-based sensing to arbitrary sensors, learnable embedding spaces, and robotic systems. With PET, I aim to bring the benefits of asynchronicity and sparsity to these new domains.

Please reach out to collaborate!