Facial expression recognition based on dimension model using sparse coding

We present an expression recognition system based on dimension model of internal states that is capable of identifying the various emotions using automated feature extraction. Feature vectors for facial expressions are extracted from a hybrid approach using fuzzy c-mean clustering algorithm and dynamic linking based on Gabor wavelet representation. The result of facial expression recognition is compared with dimensional values of internal states derived from semantic ratings of words related to emotion by experimental subjects. The dimensional model recognizes not only six facial expressions related to six basic emotions (happiness, sadness, surprise, angry, fear, disgust), but also expressions of various internal states. In this paper, with dimension model we have improved the limitation of expression recognition based on basic emotions, and have extracted features automatically with a new approach using FCM algorithm and the dynamic linking model.