Abstract

For the traditional Bayesian face recognition methods, a sim- ple prior on face representation cannot cover large variations in facial poses, illuminations, expressions, aging, and occlusions in the wild. In this paper, we propose a new approach to learn the face prior for Bayesian face recognition. First, we extend Manifold Relevance Determination to learn the identity subspace for each individual automatically. Based on the structure of the learned identity subspaces, we then propose to esti- mate Gaussian mixture densities in the observation space with Gaussian process regression. During the training of our approach, the leave-set- out algorithm is also developed for overfitting avoidance. On extensive experimental evaluations, the learned face prior can improve the per- formance of the traditional Bayesian face and other related methods significantly. It is also proved that the simple Bayesian face method with the learned face prior can handle the complex intra-personal variations such as large poses and large occlusions. Experiments on the challeng- ing LFW benchmark shows that our algorithm outperforms most of the state-of-art methods.