Bayesian Face Recognition with Deformable Image Models

We propose a novel representation for characterizing image differences using a deformable technique for obtaining pixel-wise correspondences. This representation, which is based on a deformable 3D mesh in XYI-space, is then experimentally compared with two related correspondence methods: optical flow and intensity differences. Furthermore, we make use of a probabilistic similarity measure for direct image matching based on a Bayesian analysis of image variations. We model two classes of variation in facial appearance: intra-personal and extra-personal. The probability density function for each class is estimated from training data and used to compute a similarity measure based on the a posteriori probabilities. The performance advantage of our deformable probabilistic matching technique is demonstrated using 1700 faces from the US Army’s “FERET” face database. International Conference on Image Analysis & Processing (ICIAP’01), Palermo, Italy, September 2001 This work may not be copied or reproduced in whole or in part for any commercial purpose. Permission to copy in whole or in part without payment of fee is granted for nonprofit educational and research purposes provided that all such whole or partial copies include the following: a notice that such copying is by permission of Mitsubishi Electric Research Laboratories, Inc.; an acknowledgment of the authors and individual contributions to the work; and all applicable portions of the copyright notice. Copying, reproduction, or republishing for any other purpose shall require a license with payment of fee to Mitsubishi Electric Research Laboratories, Inc. All rights reserved. Copyright c ©Mitsubishi Electric Research Laboratories, Inc., 2001 201 Broadway, Cambridge, Massachusetts 02139 ∗ MERL Research Laboratory † INRIA Rocquencourt ‡ MIT Media Laboratory Published in: International Conference on Image Analysis & Processing, (ICIAP’01), 2001.