Face Recognition Based on the Probability Support Vector Machines
Xiaoling Xiao, Li Layuan · 2008
Face recognition is very important in intelligent meeting scenario. An approach to face recognition based on the probability support vector machines is proposed in this paper. In this approach, an approach of the posterior probability output of multi-class SVM is modeled. One-against-one multi-class SVMs with probability output are chosen as the classifiers for face recognition. Considering the real need of face recognition in intelligent meeting scene, the frontal faces are detected and extracted based on the ratio between the face area and head area. The input of the SVM classifier is the raw image of the frontal face area, and the output is the probability of the tested face in each class, and the recognized face denotes the class with the highest probability. Experimental results have showed that this approach makes better application in real meeting scenario.