Using the visual Words based on Affine-SIFT descriptors for face recognition | IEEE Conference Publication | IEEE Xplore