Convex Color Image Segmentation with Optimal Transport Distances
This work addresses image segmentation for computer vision applications, but it appears incremental as it builds on existing optimal-transport frameworks without introducing major new paradigms.
The paper tackled the problem of convex, histogram-based image segmentation by using regularized optimal-transport distances as discrepancy measures, and it solved the resulting convex optimization problem with a primal-dual algorithm.
This work is about the use of regularized optimal-transport distances for convex, histogram-based image segmentation. In the considered framework, fixed exemplar histograms define a prior on the statistical features of the two regions in competition. In this paper, we investigate the use of various transport-based cost functions as discrepancy measures and rely on a primal-dual algorithm to solve the obtained convex optimization problem.