Real-time gaze-tracking embedded-system

This paper proposes a video-based Real-Time Gaze-Tracking system which allows to identify gaze commands to drive an electric powered wheelchair. The main aspect of the method is its ability to be implemented in a portable computing system, reduced both in processing capacity and in RAM memory. The Gaze-Tracking system is based on the Hough Transform, the use of shape features and the dark pupil effect. The method performance was validated using an own database, built under different ambient lighting conditions.