An Average Facial Color Image Avatar System for the Analysis by Synthesis of Affect Display by Dynamic Facial Color and Expression
Takashi Yamada · TRANSACTIONS OF THE JAPAN SOCIETY OF MECHANICAL ENGINEERS Series C · 2012
In this study, we design an average facial color image avatar called FaceAvatar, and develop the system for the analysis by synthesis of affect display by dynamic facial color and expression. More specifically, FaceAvatar is designed by applying texture mapping to the average facial color image in a 3D wire-frame model that estimates various feature points on the facial surface of the average facial color image. The developed system enables laughing operation of FaceAvatar through the use of a computer keyboard. Further, the proposed dynamic model of facial color and expression under laughing emotion are analyzed synthetically by using the system. The effectiveness of the system for the analysis by synthesis is demonstrated by sensory evaluation.