Vincent is a PhD student in the Artificial Visual Intelligence Group at the Photogrammetry and Remote Sensing Lab (AVI@PRS), ETH Zurich.
He is interested in 3D computer vision for robotics, particularly in learning powerful 3D priors about our world from ubiquitous, unannotated data using self-supervised learning. If you have an idea on how to train a "3D foundation model" from monocular videos (e.g. from YouTube), let him know.
Vincent previously completed a MSc Robotics and a BSc Electrical Engineering & Information Technology at ETH, and worked at a computer vision startup for one year. He joined TRACE in October 2025.
