A novel non-intrusive eye gaze estimation using cross-ratio under large head motion
Computer Vision and Image Understanding - Special issue on eye detection and tracking
| Hi-index | 0.00 |
When a person instructs operations to a robot, or performs a cooperative task with a robot, it is necessary to inform the robot of the person's intention and attention. Since the motion of human's face and gaze is deeply related with human's intention and attention, detection of such motion can be utilized as a natural way of communication for human-robot interaction. In this paper, we describe our real-time face tracking system based on the stereo tracking of facial feature templates and the weighted 3D model fitting. We adopted a PC-based fast vision processing system together with a field multiplexing device for stereo video signals, and realized (1) non-contact, (2) passive, (3) real-time, (4) robust, (5) compact and (6) accurate measurement system for human's head motion.