A computer mouse provides an interface between the computer and a user. The location of the pointer displayed on the screen is obtained by the detection of the two-dimensional motion with reference to the surface in contact. In its early generations, the mouse had wheels to track the movement. The wheels were later replaced by track ball and now, by light-emitting devices (LEDs) to improve its detection sensitivity and ergonomics. Despite the changes in detection hardware, the basic principle and appearance of a mouse have almost remained unchanged.
Many technologies have been demonstrated to revolutionise the tracking device, of which eye tracking is one of the most researched. Eye tracking is likely to especially benefit disabled persons who are incapable of using a mouse. In the early 1900s, simple eye tracking technology was studied using photographic films. The researchers captured the reflected light from a subject's eyes and studied this information to understand how people read and look at pictures. In recent times, researchers have employed high-resolution cameras and various types of infrared LEDs for such studies. Typically, the cameras track the movement of the pupils and the reflection of the infrared light off the cornea. The cornea is used as a reference point as it is not affected by the eye movements.
Nevertheless, eye tracking technology has many challenges. Firstly, the eyes are always in motion and it is difficult to derive intention from the constant movement. Secondly, people have different ways of looking at a computer screen and a special interface between the computer and the user is needed to avoid overloading the captured signal with unnecessary information and also to provide comfort to the user.
A researcher from Stanford University has however proposed a better eye tracking technology that could replace the old computer mouse. Coincidentally, the computer mouse was invented by the Stanford Research Institute in 1964. The researcher, Manu Kumar from the Human Computer Interaction Group has created a software and a special computer screen fitted with high-definition camera and infra-red LEDs. The core of the device is software called EyePoint that interfaces with standard eye-tracking hardware. Instead of conventional systems, Kumar decided to invoke the use of a trigger event to kick-start the eye detection process, for example the press of a button. The software requires the user to look at a Web link, for instance, and hold a 'hot key' on the keyboard while looking. By doing so, the area of the screen that is detected is enlarged and, the user could pinpoint his/her focus within the magnified region and release the hot key, effectively clicking through to the link with a quick blink.
Another unique feature is the absence of an indicator or a cursor. According to Kumar, studies have shown that a user is likely to be distracted as he or she tries to control a location cursor. In his experiment with 20 users, he observed a drop in people's performance when he used a blue dot to indicate their eyes' position. Although a high error rate of 20% was reported in the trial, about 90% of the participants indicated their preference for EyePoint. While Kumar continues to work on algorithms to improve the performance of EyePoint, he commented that EyePoint might not work well for people with thick glasses, special contact lenses, or suffering from lazy eyes. Currently, Kumar is developing more software under a project called Gaze-enhanced User Interface Design (GUIDe); one of them, called EyeExposé, is designed for Apple's operating system.
Nowadays, higher-end laptops are equipped with cameras for videoconferencing and other applications. If high-definition cameras can find their way to future laptops, it is very likely that eye tracking could be implemented as well.
For more information contact Patrick Cairns, Frost & Sullivan Technical Insights, +27 (0)21 680 3274, [email protected]
© Technews Publishing (Pty) Ltd | All Rights Reserved
printer friendly version