Since the advent of the personal computer, we have been accustomed to the keyboard and mouse as the conventional input peripherals by which we control our devices.
However, recent years have seen the development and deployment of a new technology that is set to revolutionise personal computing altogether. Whether you would like to draft an e-mail or browse through your news feed on Facebook, imagine setting your keyboard and mouse aside and instead carrying out the tasks on your computer using your eye movements alone.
This is not science fiction – rather, eye-gaze tracking is an area of very active research that has generated interest internationally. The University of Malta in collaboration with the Education Department is contributing to the development of eye-gaze tracking through a project called Eye-Communicate, which is funded by the Malta Council for Science and Technology through the National Research & Innovation Programme (2012). The aim of this project is to develop a low-cost non-intrusive eye-gaze tracking platform.
Eye-gaze tracking technology finds uses in many applications, in particular as an alternative communication channel for individuals with limited motor abilities, allowing them to operate assistive computer software using their eye movements alone.
A typical commercially available eye-gaze tracking device is equipped with a camera and a set of infrared illuminators, and is usually mounted beneath a monitor screen in front of the user. The infrared illumination is shone onto the eyes to create glints that reflect off the surface of the eyeball, and the eye-gaze direction is estimated through a mathematical process that considers the positions of these glints. However, the use of infrared illumination reduces the usability of the tracking device to typical indoor environments away from infrared sources, such as the sun. Moreover, these commercial eye-gaze trackers tend to have a relatively high price, forcing individuals who may benefit from this technology to consider alternative options.
A final stage transforms this gaze direction into screen coordinates, hence enabling the mouse cursor to move according to the movement of the eyes
Clearly, renouncing to a technological device that promises to improve your quality of life should not be an option. Hence, over the past year the Department of Systems and Control Engineering at the University of Malta has teamed up with the School Resources Division of the Education Department in a nationally funded research and development project in order to develop a low-cost eye-gaze tracking platform that is based on standard hardware but can track the eye-gaze effectively and robustly.
Relying on a simple webcam and removing specialised illumination sources, such as the infrared sources used in many commercial eye-gaze trackers, from the equation, the eye-gaze tracking platform that is presently being developed through this collaboration is based on advanced mathematical computer algorithms to track the eye-gaze. By relying on standard hardware alone, costs may be reduced and portability increased.
A first phase prototype that has been developed during the past year is capable of tracking eye movements and head motion – in this way, the user is free to move naturally during tracking. Upon the acquisition of a stream of image frames through the webcam, our prototype applies image processing techniques to locate and distinguish between the different facial features, such as eyes, nose tip and mouth. The relative positions of these facial features allow an estimation of the head pose, while the eyeball orientation is estimated after the iris centre coordinates have been extracted. The head pose and the eyeball orientation are then combined to estimate the eye-gaze direction of the user. A final stage transforms this gaze direction into screen coordinates, hence enabling the mouse cursor to move according to the movement of the eyes.
With the development of this first phase prototype, we have achieved an important milestone in this project. Ongoing work in collaboration with the Education Department during the second stage of the project will seek to improve further upon the work that has been carried out so far.
In the coming months, trained personnel from the Education Department shall be assessing the use of a commercial eye-gaze tracking device by a selected group of school children who may benefit from this technology. This study should allow us to identify specific challenges associated with eye-gaze tracking in real-life situations. This information will be used to tailor the further development of our prototype to increase robustness and ease-of-use in different environments. The same group of selected school children will also later test the developed prototype to permit an evaluation in real life.
Despite the challenging nature of this project, we are encouraged by the successful results achieved so far and motivated by a desire to make this technology accessible to everyone, especially to those individuals whose lives could be improved.