A hybrid learning approach named confusioncrossed support vector machine tree (CSVMT) has been proposed in our current work. It is developed to achieve a better performance for complex distribution problems even when the two parameters of SVM are not appropriately selected. In this paper a facial expression recognition approach based on CSVMT is proposed. Pseudo-Zernike moments are applied in the feature extraction phase, and then CSVMT learning model is performed during the facial expression recognition phase. The compared results on Cohn- Kanade facial expression database show that the proposed approach appeared higher recognition accuracy than the other approaches.
Citation:
Qinzhen Xu, Pinzheng Zhang, Wenjiang Pei, Luxi Yang, Zhenya He, "A Facial Expression Recognition Approach Based on Confusion-Crossed Support Vector Machine Tree," iih-msp, pp.309-312, 2006 International Conference on Intelligent Information Hiding and Multimedia Signal Processing (IIH-MSP'06), 2006