Vitaliy Kurlin (6/2/20): Persistence-based skeletonization of images скачать в хорошем качестве

Vitaliy Kurlin (6/2/20): Persistence-based skeletonization of images

Title: Persistence-based skeletonization of images Abstract: We consider the problem of splitting an image into a small number of convex polygons with vertices at subpixel resolution. Edges of resulting superpixels can have any direction and should adhere well to object boundaries. We discuss several recent algorithms based on a topological persistence. The first algorithm introduces a Convex Constrained Mesh (CCM) that accepts any straight line segments (or a straight-line graph based on a topological persistence) and outputs a complete mesh of convex polygons without small angles and with approximation guarantees for the given lines, joint with J.Forsythe in J Electronic Imaging, 26(6), 061609, 2017. The second algorithm builds a Persistent Resolution-Independent Mesh (PRIM) on straight-line edges that are automatically detected by their total strength of contrast differences, joint with G.Muszynski in Proceedings of CTIC 2019 (Computational Topology in Image Context).