Bayesian Face Revisited: A Joint Formulation

Dong Chen¹, Xudong Cao³, Liwei Wang², Fang Wen³, and Jian Sun³

¹University of Science and Technology of China, China
chendong@mail.ustc.edu.cn

²The Chinese University of Hong Kong Hong Kong
lwwang@cse.cuhk.edu.hk

³Microsoft Research Asia, Beijing, China
xudongca@microsoft.com
fangwen@microsoft.com
jiansun@microsoft.com

Abstract. In this paper, we revisit the classical Bayesian face recognition method by Baback Moghaddam et al. and propose a new joint formulation. The classical Bayesian method models the appearance difference between two faces. We observe that this “difference” formulation may reduce the separability between classes. Instead, we model two faces jointly with an appropriate prior on the face representation. Our joint formulation leads to an EM-like model learning at the training time and an efficient, closed-formed computation at the test time. On extensive experimental evaluations, our method is superior to the classical Bayesian face and many other supervised approaches. Our method achieved 92.4% test accuracy on the challenging Labeled Face in Wild (LFW) dataset. Comparing with current best commercial system, we reduced the error rate by 10%.

LNCS 7574, p. 566 ff.

Full article in PDF | BibTeX

lncs@springer.com
© Springer-Verlag Berlin Heidelberg 2012