Special Session 157: Advances in PDE-Based and Data-Driven Approaches for Applied Sciences

Neural network parametrized level sets for image segmentation
Cong Shi
University of Vienna
Austria
Co-Author(s):    Otmar Scherzer and Thi Lan Nhi Vu
Abstract:
The Chan-Vese functionals have proven to by a first-class method for segmentation and classification. Previously they have been implemented with level-set methods based on a pixel-wise representation of the level-sets. Later parametrized level-set approximations, such as splines, have been studied. In this talk we consider neural networks as parametrized approximations of level-set functions. We show in particular, that parametrized two-layer networks are most efficient to approximate polyhedral segments and classes. We also prove the efficiency for segmentation and classification.