Facial Expression Recognition Based on Cortex-Like Mechanisms

As an important aspect of human-computer interaction, auto facial expression recognition tries to make a computer extract and analyze human's expression from input face image. In this paper, a facial expression recognition method based on cortex-like mechanisms and visual receptive field is presented. Some biological vision theory such as excited field and inhibitory field are made use to imitate human visual perception mechanism for features extracting. The support vector machine is used for the facial expression classification. As this new feature has very good expression ability for visual perception, the experiments on JAFFE and TFEID databases indicates that the method has the better performance in robustness and accuracy than some present algorithms based on PCA, LDA, LBP and so on.