A ROBUST FEATURE EXTRACTION FRAMEWORK FOR FACE RECOGNITION (TA-P2)

Author(s) : Guang Dai (College of Computer Science, Zhejiang University, China) Yuntao Qian (College of Computer Science, Zhejiang University, China)

Abstract : This paper describes a kernel fractional-step nonlinear discrimi-nant analysis (KF-NDA) method to extract the nonlinear features. It not only extends the fractional-step linear discriminant analysis (F-LDA) method to a nonlinear version, but also further improves the generalization ability of traditional kernel non-linear discriminant analysis (K-NDA). On the other hand, the Gabor transformed face images exhibit strong characteristics of spatial locality, scale and orientation selectivity, similar to those displayed by Gabor wavelets. Such characteristics produce salient local features that are most suitable for FR. Hence, the augmented Gabor feature vector (AGFV) derived from a set of downsampled Gabor wavelet representations of face images is robust to the various of face images, and simultaneously exhibit the more discriminatory information. Based on the AGFV and the KF-NDA, a robust feature extraction framework, i.e., the Gabor KF-NDA (GKF-NDA), is proposed for face recognition (FR). In this framework, the KF-NDA method is applied to extract the robust nonlinear feature from the AGFV. Experimen-tal results tested on the popular databases show that the GKF-NDA is more effective than other existing FR approaches.

Menu