Abstract: In recent years, facial expression recognition has become a hot research direction in human computer interaction,machine learning,intelligent control and image processing. According to feature extraction and feature classification, recent developments of facial expression recognition were presented. From static images and image sequences, the methods of feature extraction can be divided into two categories. The former includes holistic methods and local methods, the later includes templatcbased methods, geometry-based methods and optical flow methods. In the classifier design, the main methods of feature classification can be categorized by Bayesian Network methods and Distance Metric methods. Finally, combining the latest productions and applications at home and abroad, the expectation of the development of facial expression recognition was given.

Key words: Facial aexpression recognition, Feature extraction, Feature classification