[attachment=5661]
Fast Recognition of Multi-view Faces with Feature Selection

Zhi-Gang Fan and Bao-Liang Lu
Departmart of Computer Science and Engineering
Shanghai Jiao Tong University, Shanghai 200030, China

---

We propose a discriminative feature selection method
utilizing support vector machines for the challenging task
of multi-view face recognition. According to the statisti-
cal relationship between the two tasks, feature selection and
multi-class classification, we integrate the two tasks into a
single consistent framework and effectively realize the goal
of discriminative feature selection. The classification pro-
cess can be made faster without degrading the generaliza-
tion performance through this discriminative feature selec-
tion method. On the UMIST multi-view face database, our
experiments show that this discriminative feature selection
method can speed up the multi-view face recognition pro-
traditional kernel subspace methods.