Making discriminative common vectors applicable to face recognition with one training image per person

Though discriminant common vector (DCV) method has obtained some success in face recognition task, it fails when only one training image per person is available. In this paper, we propose an approach to make DCV method applicable when each person has one training image. Our approach is based on the assumption that human faces share similar intrapersonal variation. The intrapersonal variation of the training set can be estimated from the collected generic face set. The proposed method was compared with PCA, E(PC)2A and SVD perturbation algorithm, and experimental results on the subset of FERET face database show the promising performance of the proposed method.