Facial expression recognition based on ASM and Multi-Instance Boosting

Shaoping Zhu

COMPUTER MODELLING & NEW TECHNOLOGIES 2014 18(12B) 323-330

Department of Information Management, Hunan University of Finance and Economics, 410205, China

In this paper, a novel method for facial expression recognition in dynamic facial images is proposed, which includes two stages of feature extraction and facial expression recognition. Firstly, Active Shape Model (ASM) is used to extract the local texture feature, and optical flow technique is determined facial velocity information, which is used to charaterize facial expression. Then, fusing the local texture feature and facial velocity information get the hybrid characteristics. Finally, Multi-Instance Boosting model is used to recognize facial expression from video sequences. In order to be learned quickly and complete the recognition, the class label information was used for the learning of the Multi-Instance Boosting model. Experiments were performed in the JAFFE database to evaluate the proposed method. The proposed method shows substantially higher accuracy at facial expression recognition than has been previously achieved and gets a recognition accuracy of 95.3%, which validates its effectiveness and meets the requirements of stable, reliable, high precision and anti-interference ability etc.