Gabor Features Based Method Using HDR (G-HDR) for Multiview Face Recognition
This paper introduces a novel algorithm named G-HDR, which is a Gabor features based method using Hierarchical Discriminant Regression (HDR) for multiview face recognition. Gabor features help to eliminate the influences to faces such as changes in illumination directions and expressions; Modified HDR tree help to get a more precise classify tree to realize the coarse-to-fine retrieval process. The most challenging things in face recognition are the illumination variation problem and the pose variation problem. The goal of Our G-HDR is to overcome both difficulties. We conducted experiments on the UMIST database and Volker Blanz’s database and got good results.
Unable to display preview. Download preview PDF.
- 1.Hwang, W.-S., Weng, J.: Hierarchical Discriminant Regression. IEEE Transactions on Pattern Analysis and Machine Intelligence 22(11) (2000)Google Scholar
- 3.Fazel-Rezai, R., Kinsner, W.: Image analysis and reconstruction using complex Gabor wavelets. In: Canadian Conference on Electrical and Computer Engineering2000, vol. 1, pp. 440–444 (2000)Google Scholar
- 4.Tao, L., Kwan, H.K.: Real-valued discrete Gabor transform for image representation. In: IEEE International Symposium on Circuits and Systems. ISCAS 2001., vol. 2, pp. 589–592 (2001)Google Scholar
- 6.Blanz, V., Vetter, T.: Face Recognition Based on 3D Shape Estimation from Single Images. Computer Graphics Technical Report No.2, University of Freiburg (2002)Google Scholar
- 8.Graham, D.B., Allinson, N.M.: In: Wechsler, H., Phillips, P.J., Bruce, V., Fogelman-Soulie, F., Huang, T.S. (eds.) Face Recognition: From Theory to Applications. NATO ASI Series F. Computer and Systems Sciences, vol. 163, pp. 446–456 (1998)Google Scholar