A system for multimodal interaction with Kinect-enabled virtual windows

Abstract

Commercial off-the-shelf gaming devices (e.g. such as Kinect) are demonstrating to have a great potential beyond their initial service purpose. In particul ar, when integrated within the environment or as part of smart objects, peripheral COTS for gaming may facilitate the definition of novel interaction methods, particularly applicable to smart spaces service concepts. In this direction, this paper describes a system prototype that makes possible to deliver multimodal interaction with the media contents in a Virtual Window. Using a Kinect device, the Interactive Window itself adjusts the video clipping to the real time perspective of the user ? who can freely move within the sensor coverage are. On the clipped video, the user is able to select objects by pointing at meaningful image sections and to initiate actions related to them. Voice orders may also complete the interaction when necessary. Although implemented for smart spaces, the service concept can also be applied to learning, remote control processes or teleconference.