Real-time Expression recognition System using Active Appearance Model and EFM

We present a continuous facial expression recognition system based on active appearance model (AAM) and enhanced fisher-discriminant model (EFM). AAM has been widely used in face tracking, face recognition, and object recognition tasks. In this study, we have implemented an independent AAM using inverse compositional image alignment method, which is a useful for the real-time system, because of its fast performance. The evaluation of this system carried out with the standard Cohn-Kanade facial expression database