Eye-Gaze Tracking Research Based on Image Processing

This paper presents an eye-gaze tracking system based on the image processing. All the computations are performed in software and the system just needs a PC camera attached to the user´s computer. We first extract the facial regions form the images using the skin-color model and connected-component analysis. Then the eye regions are detected by employing the rules and area segmentation. After the detection of eye movements, the tracking is performed using the mean-shift algorithms. The experimental results verify the feasibility and validity of the proposed eye-gaze tracking system, while there is still room for the improvement to be made.