Support Vector Clustering of Facial Expression Features

Facial expression recognition is an active research area that finds a potential application in human emotion analysis. This work presents an efficient approach of facial expression features clustering based on Support Vector Clustering (SVC). Common approaches to facial expression features clustering are designed considering two main parts: (1) features extraction, and (2) features clustering. In the process of facial expression extraction, we use Gabor features can reduce data dimensional, then we tune the parameters that define the Gaussian kernel width generator for clustering. Experiments on facial expression database have shown that these methods are effective to achieve facial expression features clustering.