Facial Expression Recognition System: A Practical Implementation
Kamal Kumar Ranga, Arpita Nagpal · 2012
Facial expression is one of the most powerful and immediate means for human beings to communicate their emotions, intentions, and opinions to each other. Facial expressions also provide information about cognitive state, such as interest, boredom, confusion, and stress. Facial expressions are natural and can express emotions sooner than people verbalize their feelings. It conveys non-verbal cues, which play an important role in interpersonal relations. Facial expressions recognition technology helps in designing intelligent human computer interfaces. In this paper facial expression recognition technique has been performed on the Indian faces extracted from a video. Initially, a live video of Indian college students is given as input to Haar classifier which traces out the faces from it. Then 42 facial feature points are detected using Active Appearance Model (AAM) technique using which we extract the facial features that are to be mapped on the extracted faces. In the last step four primary facial expressions (happy, sad, surprise, angry) have been classified using the technique support vector machine (SVM). It was very a challenging task to integrate these techniques of artificial intelligence and obtain a reasonable performance. The facial expressions recognizer proposed