Real-time facial expression recognition in the interactive game based on embedded hidden Markov model

Combine the facial expression of the game player with the game role and the context can highly enhance the interaction and attractiveness of the interactive computer game. We propose a framework to real-time facial expression recognition in the interactive computer game environment. There are two main contributions of this work. First, an embedded HMM for face recognition as stated in A. V. Nefian and M. H. Hayes III (1999) is used in real-time facial expression recognition. Second, face alignment is used in facial expression environment. In this paper, the embedded HMM uses two-dimensional discrete cosine transform (2D-DCT) coefficients as the observation vectors opposite to previous HMM approaches which use pixel intensities to form the observation vectors. Our proposed system reduces the complexity of the training and recognition system.