On modeling variations for face authentication

We present a scheme for face authentication in the presence of variations. To deal with variations, such as facial expressions and registration errors, with which traditional appearance-based methods do not perform well, we propose the eigenflow approach. In this approach, the optical flow and the optical flow residue between a test image and a training image are computed first. The optical flow is then fitted to a model that is pre-trained by applying principal component analysis (PCA) to optical flows resulting from variations caused by facial expressions and registration errors. The eigenflow residue, optimally combined with the optical flow residue using linear discriminant analysis (LDA), determines the authenticity of the test image. Experimental results show that the proposed scheme outperforms the traditional methods in the presence of expression variations and registration errors. The approach can be extended to model lighting and pose variations as well.