Weakly convex coupling continuous cuts and shape priors

Abstract

We introduce a novel approach to variational image segmentation with shape priors. Key properties are convexity of the joint energy functional and weak coupling of convex models from different domains by mapping corresponding solutions to a common space. Specifically, we combine total variation based continuous cuts for image segmentation and convex relaxations of Markov Random Field based shape priors learned from shape databases. A convergent algorithm amenable to large-scale convex programming is presented. Numerical experiments demonstrate promising synergistic performance of convex continuous cuts and convex variational shape priors under image distortions related to noise, occlusions and clutter. © 2012 Springer-Verlag.