Skip to content. Skip to navigation
CIM Menus
 

Recent Results in Graph Cuts: Illumination-Invariant Tracking and k-Pixel Interactions


Dr. Daniel Freedman < freedman@cs.rpi.edu >
Rensselaer Polytechnic Institute

February 28, 2005 at  3:00 AM
Zames Seminar Room - MC437

Over the last few years, techniques based on combinatorial optimization have gained a following in computer vision. In particular, a variety of vision algorithms have been formulated which rely on min-cut / max-flow methods. In this talk, we introduce two new results in this vein.

First, we show an application of graph-cut methods to the problem of tracking under large variations in illumination. This is a challenging problem which is relevant for a number of surveillance and military applications. Second, we present new results on sufficient conditions for min-cut optimization of energy functions with k-wise interactions of pixels. These conditions should prove useful in a variety of MRF-style problems.


Bio:
Daniel Freedman received the AB degree in physics (Magna Cum Laude, Phi Beta Kappa, Sigma Xi) from Princeton University in 1993 and the PhD degree in engineering sciences from Harvard University in 2000. He has been at Rensselaer Polytechnic Institute (RPI) in Troy, New York since 2000, where he is currently an assistant professor in the Department of Computer Science.

His general research interests are in Computer Vision and Geometric Algorithms. Within the field of Computer Vision, he is interested in tracking, segmentation,learning, partial differential equations, and graph cuts, with applications in medicine and military scenarios. His research in Geometric Algorithms focuses on theoretical problems concerning sampling of manifolds and the characterization of shape-space using ideas from algebraic topology.