GEMINI: Gestural multimodal interaction for intelligent networked building environments

点击翻译按钮获取中文摘要

Modern networking technologies provide new powerful solutions for the implementation of a large variety ofhuman-centered ubiquitous computer applications in the emergent e-Society such as intelligent networkedspaces for living (smart-houses), working, health and elderly care, shopping, entertainment, etc.In order to augment the functionality and user-satisfaction of these intensive-networked applications we have todevelop more accessible and unobtrusive, user-oriented and user-friendly, human-computer interactionmodalities allowing humans to seamlessly access information anytime and anywhere and to interact naturallyand intuitively with these new intelligent spaces.One of the hottest areas of interest in intelligent buildings is digital signage. This area has a strong industrialand commercial appeal. The dynamic digital signage industry is a fast-growing industry with investmentscoming from important companies such as IBM, 3M Digital Signage, and Cisco. Using plasma and LCDscreens that are usually centrally controlled and have changing contents, digital signage systems are used foradvertisements, live television, news, way-finding and other property-related information.While digital signage is, in many ways, a mature industry, the interactive digital signage is still in its infancy.The objective of this project is to develop new computer vision and voice recognition techniques that allowmultimodal interactive capabilities to be added to networked digital signage systems through gestural (bodyposture and hand gesture) and voice human-computer interfaces.The application potential spans a variety of intelligent networked building environments such as intelligentbuildings, art galleries and museums, movie complexes, libraries, shopping centers, smart health and elderlycare facilities, smart homes, etc.