Dr. Christian Rupprecht is a postdoctoral researcher at the Visual Geometry Group with Andrea Vedaldi in Oxford. He received his PhD from the Technical University of Munich in Germany advised by Nassir Navab and Gregory D. Hager (JHU). He also spent six months with Chris Pal at the Mila Institute in Montreal working on AI safety.
He is interested in self-supervised and minimally supervised learning for computer vision. More specifically, his focus is on reducing the amount of human supervision that is needed in training neural networks for various tasks such as geometry estimation and representation learning.
Deep learning in Computer Vision works exceptionally well with copious amounts of annotated training examples. However, collecting this data is often tedious, expensive, and sometimes even infeasible.
This talk explores what we can learn from a reduced amount of annotated data and how including physical priors about the world can substitute manual supervision. We will investigate examples where incorporating explicit knowledge about the world in the model leads to more interpretable predictions and better generalization.