Christian Rupprecht Seminar at the Department of Engineering Science, University of Oxford

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.