Face Identification From One Single Sample Face Image (TA-P2)

Author(s) : Hung-Son Le (Dept. of Applied Physics and Electronics, Umea University, Sweden) Haibo Li (Dept. of Applied Physics and Electronics, Umea University, Sweden)

Abstract : This paper is addressing a challenging face recognition problem: Face Identification From One Single Face Image. We present a novel approach to face identification, which is capable to identify a person from face images that are significantly different from the sample image in terms of illumination, camera view angles and expressions. The approach is based on a new way of extracting observation sequences and a simple rule to compute the dissimilarity between two face images. A person is identified based on the smallest dissimilarity, which is the summation of the dissimilarities of all pairs of observations extracted from images on both vertical and horizontal dimensions. The Haar Wavelet transform is applied to the face image as a pre-processing to lessen the dimension of the observation vectors. Our experiment results tested on both the AR face database and CMU PIE face database show that the proposed method outperforms the PCA, LDA, LFA based approaches.

Menu