Seminars/Colloquia/Invited Talks

Seminars

Leo Grady

A general purpose image segmentation algorithm using analytically evaluated random walks

PLACE:	Clark 110
EVENT:	CIS Seminar Series
DATE:	October 2, 2007
TIME:	1:00 - 2:00 PM

Abstract

An ideal image segmentation algorithm could be applied equally to the problem of isolating organs in a medical acquisition volume or to editing a digital photograph without modifying the algorithm, changing parameters, or sacrificing segmentation quality. However, a general-purpose, multiway segmentation of objects in an image/volume remains a challenging problem. In this talk, I will describe a recently developed approach to this problem that inputs a few training points from a user (e.g., from mouse clicks) and produces a segmentation by computing the probabilities that a random walker leaving unlabeled pixels/voxels will first strike the training set. These probabilities may be computed analytically and deterministically by noting the exact mathematical equivalence with a combinatorial Laplace equation. The algorithm is developed on an arbitrary, weighted graph in order to maximize the breadth of application. The solution of a combinatorial Laplace equation on an arbitrary graph admits interpretation of the algorithm as a steady-state electrical circuit simulation. I will illustrate the use of this approach with examples from several image segmentation problems (without modifying the algorithm or the single free parameter), compare this algorithm to other approaches, discuss the theoretical properties that describe its behavior and detail some new results.

Brief Biography:

Leo Grady earned a BSc in Electrical Engineering at the University of Vermont in 1999 and the PhD in the Cognitive and Neural Systems department at Boston University in 2003. Since Autumn 2003, he has been a member of the technical staff at Siemens Corporate Research in the Imaging and Visualization Department, Princeton. His research primarily focuses on image segmentation, data clustering, learning and filtering using techniques from graph theory, combinatorial topology and partial differential equations. Other interests include pattern/object recognition, applied mathematics, non-uniform data processing, image registration, cellular automata, machine learning, robotics and emergent phenomena.