Space-variant model for image segmentation over surfacesPP

We address the segmentation of piecewise-smooth real-valued functions having values on a complete, connected, 2-manifold embedded in $\R^3$. The main idea is to consider the functions as a sum of a piecewise constant component and a smooth component and to spatially combine total variation and L2 regularizations. The formulation is based on the minimization of a Convex Non-Convex functional where an ad-hoc non-convex regularization term improves the treatment of the boundary of the segmented regions.

This is poster number 30 in Poster Session

Martin Huska (University of Bologna)
computer graphics, image segmentation, images on manifold, manifold segmentation