Evaluation of a Low-cost Eye Tracking System for Computer Input

Abstract

Eye tracking technology has become one of the most popular techniques within the human and computer interaction (HCI) era, this is especially important research for people which have difficulty with speech and movement disabilities. The primary function of this technology is based on a device that tracks the movement of the eye to identify a position or scan a display. Suitable devices for eye movement can then be integrated in concordance with the requirements of the organization. Currently, eye tracking devices are becoming increasingly cheaper which make them an interesting resource for research. Although numerous studies have been conducted involving applications of eye tracking with a low-cost device, few studies have compared the actual eye tracking systems themselves. This paper empirically evaluated the performance of three independent low-cost eye tracking devices, Gazepoint’s GP3, EyeTribe, and DIY (Do-It-Yourselt). The performance evaluation used a multi-directional point-select task based on visual searching and selection to change the ground colour of a circular menu conforming to ISO 9241-9 standards for computer pointing devices. Results indicated that the spatial accuracy and speed were a good reflection of targeted tracking of errors, completion time and throughput. Consequently, the experiment with the target task showed that all devices can be a potentially valuable resource for human computer interaction research. Additionally, this basic result will be used to develop an advanced system for Thai text entry to aid communication for handicapped users.