Development and evaluation of low-cost eye tracking performance

Abstract

Nowadays, Eye Tracking Technology is one main challenge for developers of new human-computer interface. This technology provides a more natural way of interacting with a computer system by user’s eye. Now commercially available systems exist; however, they are not widely used because of expensiveness. This paper describes a tutorial that shows step by step how to build a low-cost eye tracker using eyeglass setup. The glasses are designed to work with video-oculography technique (VOG) in the infrared spectrum. The key features of the eye is a center of the pupil. The gaze point of users are estimated based on order polynomial regression (linear, quadratic and cubic). This research conducted an experiment with 30 participants tested eye-gaze calibration. Consequently, the experimental results showed that eye-gaze can be estimated in real-time, while the cubic function can be accurate eye-gaze position more than the other function.