Introduces a set of effective computational techniques forunderstanding the visual aspects of human interaction, which could beused, for example, in videoconferencing applications. First, we presentmethods for face detection and the capture of attentional frames tofocus the processing for visually mediated interaction. Second, wepresent methods for recognising the various gesture phases that can beused to control the camera systems in the integrated system. Finally, wediscuss how these techniques can be extended to “virtualgroups” of multiple people interacting at multiple sites
展开▼