Improving kernel Fisher discriminant analysis for face recognition

@article{citeulike:2783273, abstract = {This work is a continuation and extension of our previous research where kernel Fisher discriminant analysis (KFDA), a combination of the kernel trick with Fisher linear discriminant analysis (FLDA), was introduced to represent facial features for face recognition. This work makes three main contributions to further improving the performance of KFDA. First, a new kernel function, called the cosine kernel, is proposed to increase the discriminating capability of the original polynomial kernel function. Second, a geometry-based feature vector selection scheme is adopted to reduce the computational complexity of KFDA. Third, a variant of the nearest feature line classifier is employed to enhance the recognition performance further as it can produce virtual samples to make up for the shortage of training samples. Experiments have been carried out on a mixed database with 125 persons and 970 images and they demonstrate the effectiveness of the improvements.}, author = {Liu, Qingshan and Lu, Hanqing and Ma, Songde }, booktitle = {Circuits and Systems for Video Technology, IEEE Transactions on}, citeulike-article-id = {2783273}, doi = {10.1109/TCSVT.2003.818352}, journal = {Circuits and Systems for Video Technology, IEEE Transactions on}, keywords = {kernels}, number = {1}, pages = {42--49}, posted-at = {2008-05-10 21:37:46}, priority = {2}, title = {Improving kernel Fisher discriminant analysis for face recognition}, url = {http://dx.doi.org/10.1109/TCSVT.2003.818352}, volume = {14}, year = {2004} }

TY - JOUR ID - citeulike:2783273 L3 - citeulike-article-id:2783273 TI - Improving kernel Fisher discriminant analysis for face recognition T2 - Circuits and Systems for Video Technology, IEEE Transactions on JF - Circuits and Systems for Video Technology, IEEE Transactions on VL - 14 IS - 1 SP - 42 EP - 49 N2 - This work is a continuation and extension of our previous research where kernel Fisher discriminant analysis (KFDA), a combination of the kernel trick with Fisher linear discriminant analysis (FLDA), was introduced to represent facial features for face recognition. This work makes three main contributions to further improving the performance of KFDA. First, a new kernel function, called the cosine kernel, is proposed to increase the discriminating capability of the original polynomial kernel function. Second, a geometry-based feature vector selection scheme is adopted to reduce the computational complexity of KFDA. Third, a variant of the nearest feature line classifier is employed to enhance the recognition performance further as it can produce virtual samples to make up for the shortage of training samples. Experiments have been carried out on a mixed database with 125 persons and 970 images and they demonstrate the effectiveness of the improvements. KW - kernels AU - Liu, Qingshan AU - Lu, Hanqing AU - Ma, Songde PY - 2004/// UR - http://dx.doi.org/10.1109/TCSVT.2003.818352 ER -