Abstract

We attempt to automate facial expression recognition and introduce it into the man­machine interaction as a new modality. This will make the interaction compact and more efficient. As the first step, we developed a self­adaptive expert system that accepts the facial features contours localized in a static dual­view facial image and returns the expression interpretation label(s) used by the user. Expression identification in terms of the encountered facial actions is also displayed to the user. Reasoning with uncertainty about the extracted facial expression data is employed for facial action coding and quantification. A memory of experiences, inspired by the Shank's theory of human autobiographical memory organization and the instance­based learning, expounds the encoded facial actions in terms of the learned interpretation labels. Validation studies on the prototype suggest that the expressions' identifications and interpretations achieved are generally consistent with those defined by the users.