An Experimental Approach to Tracking User Eye Movement in Interactive Apps - Implications for Design.

A gesture consists of a sequence of continuous elements, typically strokes, which are performed in a sequential time order. To detect a gesture, it is necessary to know when the gesture starts and when it ends. The problem encountered separating gaze gestures from natural eye movements is very similar to the challenge of recognizing spoken commands for computer input where a normal conversation should not trigger commands. This paper focuses on finding a reasonable alphabet which is separable from natural eye movement. With a vision of ubiquitous computing, the amount of interaction with computers will increase and we are in need of interaction techniques which do not cause physical problems. The user study that was undertaken here consists of three tasks to achieve gaze gesture. The first was to close a dialog box using gaze command instead of a mouse. The second used help lines to guide the gaze gesture while the third was to surf the Internet for 3 minutes. The paper presents a usability tool for recording of interaction and gaze activity. It also describes algorithms for reading detection. The approach presented here will release results based on user study conducted with prototypes developed for the purpose.