Subspace Linear Discriminant Analysis for Face Recognition

In this paper we describe a holistic face recognition method based on subspace Linear Dis-criminant Analysis (LDA). The method consists of two steps: rst we project the face image from the original vector space to a face subspace via Principal Component Analysis where the subspace dimension is carefully chosen, and then we use LDA to obtain a linear classiier in the subspace. The criterion we use to choose the subspace dimension enables us to generate class-separable features via LDA from the full subspace representation. Hence we are able to solve the generalization/overrtting problem when we perform face recognition on a large face dataset but with very few training face images available per testing person. In addition, we employ a weighted distance metric guided by the LDA eigenvalues to improve the performance of the subspace LDA method. Finally, the improved performance of the subspace LDA approach is demonstrated through experiments using the FERET dataset for face recognition/veriication, a large mugshot dataset for person veriication, and the MPEG-7 dataset. We believe that this approach provides a useful framework for other image recognition tasks as well.