Real-Time Facial Expression Classification
Wang Yubo Ai Haizhou Wu Bo Huang Chang (State Key Laboratory of Intelligent Technology and Systems, Department of Computer Science and Technology, Tsinghua University, Beijing 100084)
In this paper, a real-time facial expression classification method based on real Adaboost algorithm is presented. Using Haar-like features weak classifiers of look-up-table (LUT) type, that have confidences in real values as their outputs, are designed, and correspondingly by using real Adaboost algorithm facial expression classifiers are learnt. The experimental results show that, in comparison with support vector machines (SVMs), this method achieves almost the same correction rate, and is nearly 300 times faster in speed. It could be almost in real time, and is of significance in applications.