Probabilistic matching for face recognition

We propose a new technique for direct visual matching of images for the purposes of face recognition, database search and image retrieval. Specifically, we argue in favor of a probabilistic measure of similarity, in contrast to simpler methods which are based on standard L/sub 2/ norms (e.g., template matching) or subspace-restricted norms (e.g., eigenspace matching). The proposed similarity measure is based on a Bayesian analysis using two mutually-exclusive classes of image variation as encountered in a typical face recognition task. The high-dimensional probability density functions for each respective class are obtained from training data using an eigenspace density estimation technique and subsequently used to compute a similarity measure based on the relevant a posteriori probability, which is used to rank matches in the database. The performance advantage of this probabilistic matching technique over standard nearest-neighbor eigenspace matching is demonstrated using results from ARPA's 1996 "FERET" face recognition competition, in which this algorithm was found to be the top performer by a 10% (or better) margin to the other competitors.