F. Ren, D. Jian, H. Xiang, S. Kuroiwa, T. Tanioka, Z. Zhang (Japan), and C. Zong (PRC)
Recognizing human emotion, Creating machine emotion, Mental state transition network, Affective interface
Information recognition and extraction of human emotions from words and actions, facial expression and sensibil ity are necessary for machines in order to communicate smoothly with humans and to realize affective communica tion. We focus on human psychological characteristics to develop general-purpose agents that can recognize human emotions and create machine emotions. We comprehen sively analyze voice sounds and picture images that rep resent information included in emotion elements such as phonation, facial expressions, and language expressions. We analyze and estimate extensive statistical data based on the latest achievements of brain science and psychology in order to derive transition networks for human psycho logical states. We establish a speaker word model for the computer to simulate its own psychological transitions and emotional presentations, which is aimed to develop affec tive interface and establish theoretic structure and realiza tion method for emotion communication. In this paper, a Mental State Transition Network is presented and an affec tive interface based on the presented network is proposed, and some psychological experiments are given.
Important Links:
Go Back
