The use and importance of computers and other electronic devices such as mobile phones
in our daily lives has increased. As a result the need to develop more convenient ways of
interaction with these systems has gained impetus. A gaze tracker, a system capable of
detecting the line of gaze of a person, is a good example of a human-computer interaction
system. Gaze trackers win over traditional devices such as mouse and keyboard because
they provide a more intuitive interaction mean and are most beneficial to the physically
challenged.
The main issue concerning gaze trackers is their high purchase cost making their
accessibility less probable to the common consumer. This thesis addresses the main issue
in Gaze Tracking, being high system cost. The work proposes the use of low cost devices
such as web cameras and light emitting diodes to design a real-time gaze tracker for
interaction with computer screens. It includes the design and implementation of two
different approaches for real-time gaze tracking with a main focus on system cost. A full
system approach has been employed by implementing a face detection and eye detection
system to assist in gaze tracking.