Computer Vision, IEEE International Conference on (2001)
Vancouver, B.C., Canada
July 7, 2001 to July 14, 2001
Bernd Heisele , Massachusetts Institute of Technology
Purdy Ho , Massachusetts Institute of Technology
Tomaso Poggio , Massachusetts Institute of Technology
We present a component-based method and two global methods for face recognition and evaluate them with respect to robustness against pose changes. In the component system we first locate facial components, extract them and combine them into a single feature vector which is classified by a Support Vector Machine (SVM). The two global systems recognize faces by classifying a single feature vector consisting of the gray values of the whole face image. In the first global system we trained a single SVM classifier for each person in the database. The second system consists of sets of viewpoint-specific SVM classifiers and involves clustering during training. We performed extensive tests on a database which included faces rotated up to about 40° in depth. The component system clearly outperformed both global systems on all tests.
T. Poggio, B. Heisele and P. Ho, "Face Recognition with Support Vector Machines: Global versus Component-based Approach," Computer Vision, IEEE International Conference on(ICCV), Vancouver, B.C., Canada, 2001, pp. 688.