Eye Gaze Correction with Stereovision for Video-Teleconferencing
ECCV '02 Proceedings of the 7th European Conference on Computer Vision-Part II
WACV '96 Proceedings of the 3rd IEEE Workshop on Applications of Computer Vision (WACV '96)
Head Pose Estimation in Computer Vision: A Survey
IEEE Transactions on Pattern Analysis and Machine Intelligence
Movable cameras enhance social telepresence in media spaces
Proceedings of the SIGCHI Conference on Human Factors in Computing Systems
HRI '12 Proceedings of the seventh annual ACM/IEEE international conference on Human-Robot Interaction
Hi-index | 0.00 |
Video conferencing is widely used with the growing popularity of Internet call systems such as FaceTime, Skype and broad band network. However, people still find it unnatural compared to face-to-face talk. One of the problems is that the display screen and web camera are fixed and do not track a person when he/she moves, resulting in an incomplete view and unnaturalness to the other party. In this paper, we propose a more engaging telepresence solution based on a tablet mounted on a pan-tilt robotic base. During video-conferencing, the human face is tracked by the built-in camera on the tablet. The tracking results will drive the robotic base in real-time such that the display and camera will follow the moving face. In this way, it not only sends the captured face images, but also gives natural feeling of head movement as in a face-to-face conversation. We conducted user studies and the findings showed preference of the proposed system.