Monday, February 16th
3-4pm, ST2 Room 430


Bayesian Approaches to Image Segmentation

Dr. Daniel Cremers
Department of Computer Science
UCLA


Abstract:

When segmenting their environment into meaningful regions, human observers exploit a number of low-level cues (such as intensity, color, texture or motion information) and higher level knowledge about objects of interest. In my presentation, I will present ways to incorporate such information into image segmentation methods. In particular, I will present: - the 'Diffusion Snake' as a fast spline-based implementation of the Mumford-Shah functional - 'Motion Competition' as an extension of the Mumford-Shah framework from intensity segmentation to motion segmentation. Segmenting contours are represented either by splines or by level sets. - the integration of higher-level statistical shape priors into the segmentation processes. This permits to cope with noise, background clutter and partial occlusions of the objects of interest. More information can be found under http://www.cs.ucla.edu/~cremers