Decision and Control Lecture Series

Decision and Control Laboratory, Coordinated Science Laboratory

Visual Tracking with Active Contours in Sobolev Spaces

Anthony Yezzi

Byers Professor, Systems and Controls

School of Electrical and Computer Engineering, Georgia Tech

Monday, April 22, 2013

11:00 a.m. to 12:00 p.m.

141 Coordinated Science Laboratory

___________________________________________________________________________________________________________________________________________________________________________________________

Abstract

Since the introduction of "snakes" (parametric deformable contours) by Kass, Witkin, and Terzopolous, "active contours" have become popular both for still image segmentation as well as for tracking objects in image sequences (such as in video). There is a wide variety of active contour models just as there is a wide variety of techniques to incorporate them into the tracking problem. We will focus on purely geometric active contours (parameterization independent) whose evolutions are derived as the gradient descent of a meaningfully designed energy functional.

While the simplest tracking algorithm for such classes of active contours is simply to run "serial" gradient descent by using the steady state output contour of the gradient descent evolution on the previous image as the starting point for a new gradient descent on the following image, such a straight-forward technique is very sensitive to even the most momentary visual distractions (temporary occlusions, moving shadows, rapid changes between consecutive images, etc.).

Better techniques incorporate some manner of interdependence between computed contours in consecutive frames, thereby making it more difficult for tracking to be thrown off within a single image. In this talk we will present a Sobolev-type metric on the space of curves that not only improves the performance of the simplistic serial gradient descent based tracker but also allows for predictor/corrector schemes which are geometrically consistent with the gradient descent formulation.

Biography

Professor Yezzi was born in Gainsville, Florida and grew up in Minneapolis, Minnesota. He obtained both his Bachelor's degree and his Ph.D. in the Department of Electrical Engineering at the University of Minnesota with minors in mathematics and music.

After completing his Ph.D., he continued his research as a post-Doctoral Research Associate at the Laboratory for Information and Decision Systems at Massachusetts Institute of Technology in Boston, MA.

His research interests fall broadly within the fields of image processing and computer vision. In particular he is interested in curve and surface evolution theory and partial differential equation techniques as they apply to topics within these fields (such as segmentation, image smoothing and enhancement, optical flow, stereo disparity, shape from shading, object recognition, and visual tracking).

Much of Dr. Yezzi's work is particularly tailored to problems in medical imaging, including cardiac ultrasound, MRI, and CT. He joined the Georgia Tech faculty in the fall of 1999 where he has taught courses in DSP and is working to develop advanced courses in computer vision and medical image processing. Professor Yezzi consults with industry in the areas of visual inspection and medical imaging. His hobbies include classical guitar, opera, and martial arts.

PLEASE JOIN US FOR COOKIES AND COFFEE AT 10:30AM BEFORE THE SEMINAR IN ROOM 154 COORDINATED SCIENCE LABORATORY