We will review the problem of face geometric structure reconstruction from a single image, planar shape matching, and the dimensionality of facial expressions all from the point of view of deep learning. Geometry - which is in a sense understanding the problem at hand by finding its invariants will be shown to be related in some specific cases to deep learning. In this talk we will try to unify learning and understanding by empirically understanding deep learning.

This presentation is part of Minisymposium “MS29 - Geometry and Learning in 3D Shape Analysis”
organized by: Ronald Lui (Chinese University of Hong Kong) , Rongjie Lai (Rensselaer Polytechnic Institute) .

Authors:

Ron Kimmel (Technion - Israel Institute of Technology)

Keywords:

deep learning