Seminars/Colloquia/Invited Talks

Seminars

Shantanu H. Joshi

Intrinsic Bayesian Active Contours

PLACE:	Clark 314
EVENT:	CIS Seminar Series
DATE:	January 30, 2007
TIME:	1:00 - 2:00 PM

Abstract

I will present a framework for incorporating prior information about high-probability shapes in the process of contour extraction and object recognition in images. Here one studies shapes as elements of an infinite-dimensional, non-linear, quotient space, and statistics of shapes are defined and computed intrinsically using differential geometry of this shape space. Prior probability models are constructed on the tangent bundle of shape space. Past work on boundary extraction has used active curves driven by vector fields that were based on image gradients and roughness penalties. The proposed method incorporates a prior knowledge of shapes in the form of gradient fields in addition to the previously used image vector fields. Through experimental results, we demonstrate the use of prior shape models in estimation of object boundaries, and their success in handling partial obscuration and missing data. Furthermore, we describe the use of this framework in shape-based object recognition or classification. This is joint work with Anuj Srivastava, and Eric Klassen, FSU.

Brief Biography:

Joshi received a B.Eng. in Electronics and Telecommunication from the University of Pune, India in 1998, and a M.Sc. in Electrical Engineering from the Florida State University in 2002. He is currently pursuing a Ph.D. in the Department of Electrical Engineering at Florida State University and working as a research assistant in the Center for Applied Vision and Imaging Sciences at Florida State University. His current research interests include image analysis and computational differential geometry with applications to computer vision.