Real time gesture recognition using eigenspace from multi-input image sequences

The paper presents a real time gesture recognition method for interactive systems which can perform an interaction like human communication. This method recognizes a user's gesture using the eigenspace constructed from multi-input image sequences. Since this eigenspace represents the approximate 3D information of gesture, complicated gestures which may have self-occlusion and confusion in a single input image sequence can be recognized correctly. Moreover, the method is suitable for obtaining the degree information of gesture, such as speed and magnitude. Using the method, the authors realize a real time interactive system, the Virtual Conductor System, which can control music played by a computer using gesture recognition results, and indicates the usefulness of the method.