Diffusion-snakes: combining statistical shape knowledge and image information in a variational framework

Author

Cremers, Daniel ; Schnörr, Christoph ; Weickert, Joachim

Author_Institution

Dept. of Math. & Comput. Sci., Mannheim Univ., Germany

fYear

2001

fDate

2001

Firstpage

137

Lastpage

144

Abstract

We present a modification of the Mumford-Shah functional and its cartoon limit which allows the incorporation of statistical shape knowledge in a single energy functional. We show segmentation results on artificial and real-world images with and without prior shape information. In the case of occlusion and strongly cluttered background the shape prior significantly improves segmentation. Finally we compare our results to those obtained by a level-set implementation of geodesic active contours

Keywords

computational geometry; functional equations; image segmentation; statistical analysis; variational techniques; Mumford-Shah functional; cartoon limit; cluttered background; diffusion snakes; energy functional; image information; image segmentation; occlusion; statistical shape knowledge; variational framework; Active contours; Active shape model; Computer graphics; Computer vision; Equations; Image segmentation; Level set; Shape control; Spline; Statistical learning;

fLanguage

English

Publisher

ieee

Conference_Titel

Variational and Level Set Methods in Computer Vision, 2001. Proceedings. IEEE Workshop on

Conference_Location

Vancouver, BC

Print_ISBN

0-7695-1278-X

Type

conf

DOI

10.1109/VLSM.2001.938892

Filename

938892