In this talk, we will discuss the use of geometric curvature based on flows in image processing and computer vision. Certain of these flows give rise to (nonlinear) geometric diffusion equations which are invariant with respect to a given transformation group action. The smoothing properties of such flows will be considered. We will provide some relevant results from the theory of curve and surface evolution, and show how these may be used for a number of areas in computer vision and image processing, such as image enhancement, optical flow, image segmentation, shape theory, and invariant scale-spaces. We will also discuss a new paradigm for deformable contours ("snakes") which may be used for visual tracking.
firstname.lastname@example.org (612) 625-8000.