Abstract
A face includes many nonverbal and personal informations concerned with communications. It is important for us to handle a face with engineering treatment by apply to man-machine interface thinking over one's emotion or intention. In this paper, we propose a method of facial classification as simple as possible. We defined "movement patterns" based on the charactaristic of facial parts during facial expressions. We analyzed these movements, and defined the patterns of facial deformation. And, we made emotion templates for discrimination of facial expressions using defined patterns.
