Automatic emotion recognition in video

Automatic recognition of emotion is becoming increasingly important component in Human-Machine Interaction (HMI) systems. Recognizing human facial emotion by computer is interesting and challenging problem. Emotions are the key semantic component of human communication. This paper presents the system for recognizing emotions through facial expression displayed in the video. Audio-visual emotion recognition can be carried out with the video sequence. The video sequence is a mixture of both the audio and video information. But in this paper, we are dealing only with the video information to recognize the emotion of a character. This process is carried out by the selection of the target frame followed by the segmentation of the video sequence. Facial feature tracking is performed to extract features. The face is classified in to different Action Units. Face emotions are recognized by using Dynamic Bayesian Network. The results so far shows the approach to have a promising success rate.