Towards effective eye pointing for gaze-enhanced human-computer interaction

In this paper we present some solutions we have devised to improve eye pointing in ordinary PC use. While eye tracking technology has advanced considerably in the last years, several limitations prevent gaze-based interaction from becoming as comfortable and effective as traditional mouse or touchpad input. The proposed approach combines both automatic and manual error correction techniques (namely eye data filtering, virtual magnetization, magnifying glass and cursor manual shift) to allow enhanced exploitation of gaze as an alternative or additional communication channel. Experiments have shown that the approach can provide better performance than touchpad pointing and compares fairly with the mouse.