Common Vector Based Face Recognition Algorithm

Abstract

The common vector represents the common invariant properties of the respective class, which was proposed and originally introduced for isolated word recognition problems. Inspired by the idea of the common vector, researchers proposed the discriminant common vector approach and the common faces approach for face recognition. In this chapter, we study an approach for face recognition based on the difference vector plus the kernel PCA (Principal component analysis). Difference vector is the difference between the original image and the common vector which is obtained by the images processed by the Gram-Schmidt orthogonalization and represents the common invariant properties of the class. The optimal feature vectors are obtained by KPCA (kernel-based PCA) procedure for the difference vectors. Recognition result is derived from finding the minimum distance between the testing difference feature vectors and the training difference feature vectors. A comparative study among them in face recognition (or generally in the SSS problem) was carried out. To test and evaluate these approach performance, a series of experiments are performed on five face databases: ORL, YALE, FERET, AR and JAFFE face databases and the experimental results show that the common vector based face recognition algorithm is encouraging.