KANSEI analysis of dance performance

We propose a method to realize extractions of emotional information from human gestures in real time. The basic ideas of understanding the human motion is based on Rudolf Laban's theoretical work. The system we developed consists of two parts: image processing; and motion analysis. The former part works to obtain human movement by using two color CCD cameras, while the latter is involved in mapping physical parameters onto emotional information. We are trying to realize a sort of "KANSEI extractor" as the last goal. Humans can understand the emotional mode, such as happy, angry and sad, in which mode the actor performs. It seems that human movement carries emotional information although it is difficult to express the emotional effects with particular physical parameters. From this point of view, we aim to extract emotional information in dance performance with the aid of a computer system based on Laban's theory of movement. In our proposed method, human motions are mapped onto low-dimensional parameter space in real time introducing two spaces: the general space- and personal space.