In this paper, we present an application of nonlinear wavelet approximation to recognize faces. The advantages of nonlinear wavelet approximation are compared with linear wavelet approximation. An efficient scheme in applying nonlinear wavelet transform to face recognition is presented. Experiments using the Yale Face Library show very encouraging recognition results and that this method has the desirable property of being insensitive to facial expressions.
Index Terms:
face recognition, wavelet basis, nonlinear approximation
Citation:
L. H. Yang, T. D. Bui, C. Y. Suen, "An Application of Nonlinear Wavelet Approximation to Face Recognition," icpr, vol. 2, pp.20048, 16th International Conference on Pattern Recognition (ICPR'02) - Volume 2, 2002