A fast face recognition system on mobile phone

Face recognition on mobile phones is gaining increasing attention due to the great variety of applications it can offer. In this paper, a fast recognition system on mobile phone is presented. First, Haar-like features with AdaBoost is used to detect face in pictures taken from phone camera. Second, in order to detect the eyes of face, we use a fast and effective method - eyes template matching in specific regions of the face, and then the local binary pattern (LBP) is adopted for fast face recognition. The experimental results demonstrate that the system introduced has a good performance in terms of recognition time and rate. Besides, various optimization techniques to speed up the face detection and recognition process are discussed.