Relevant studies of human-computer interaction indicate that human nature is of increasing importance to the design of novel mechanisms for replacing a keyboard and mouse. Therefore, gesture recognition, which involves the analysis of meaningful and expressive motions by users, has attracted significant research interest. Note that gestures generally include movements of the fingers, hands, arms, face, head, and body. In this study, we focus on posture identification, which involves only body positions. We also address several feasible applications of posture identification, including education, training, sports, films, video games, product design and even medical diagnosis. One of several alternatives, the markerless approach to motion capture, is adopted in our proposed scheme. It is not required for users to wear special clothes or equipment. The Kinect, which is a 3D motion sensor for the Xbox 360 game console developed by Microsoft, is used to build a prototype of the proposed system. Unlike previous motion capture technologies, the Kinect is affordable and can be used in home environments. Furthermore, we propose to develop a toolkit based on the Kinect SDK for posture identification and analysis. To demonstrate the usefulness of the proposed scheme, a use case is presented and evaluated.
All Science Journal Classification (ASJC) codes
- Computer Science(all)
- Health(social science)
- Environmental Science(all)