Date of Award
There is great potential for data that might be collected by webcams and used to benefit the user. One such application of this potential is eye-gaze tracking via webcams. Knowledge about where the user is looking can be used to help the user control the computer and access data more quickly. Pupil and eye-gaze tracking can be achieved by intrusive methods (where equipment is on or inside the user) and non-intrusive methods (where equipment is apart from the user). Intrusive methods are not preferred because they are frequently cumbersome and require a large investment in equipment. One nonintrusive approach to eye-gaze tracking involves active infrared illumination. Many methods that use this approach work, but all have areas where they can be improved. This thesis study investigates a novel approach to pupil tracking that is inspired by previous research and attempts to solve some of their problems. This method proves to be functional and potentially robust. Additionally, we propose an extension of this method that can be applied to track eye-gaze and a series of base mechanics that can be used to integrate interface by gaze into contemporary user interfaces. Finally, we find that interface by eye-gaze is an achievable and worthwhile goal for the advancement of Human-Computer Interaction.
Pillai, Vijay, "A Low-Cost, Feature-Based Color Threshold Selection Approach to Eye-gaze Tracking and a Basis For Gaze Interface" (2017). Senior Theses. 1150.