         
|
GRASP Lab Seminar 2003-2004
February 17 (Tuesday), 3:00 PM, Levine Hall 307, hosted by Kostas Daniilidis.
Daniel Cremers
University of California
Bayesian Approaches to Image Segmentation
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.Biography: Dr. Daniel Cremers received Bachelor degrees in Mathematics (1994) and Physics (1994), and a Master's degree in Theoretical Physics (1997) from the University of Heidelberg. He spent the year 1994/1995 as a Fulbright scholar at the Indiana State University and at the State University of New York at Stony Brook. He spent one year (1998/99) doing research in Computational Neuroscience with Andreas Herz at the Institute of Theoretical Biology of the Humboldt University in Berlin. In 1999 he joined the Computer Vision Group of Christoph Schnorr in the Department of Mathematics and Computer Science at the University of Mannheim, where he received a Phd in 2002. Since 2002 he is with Stefano Soatto in the Department of Computer Science at the UCLA.
