If the main problem you’ve had with your smartphone is the fact that you have to use your fingers to control it, rejoice: the era of eye-tracking control systems is officially here, courtesy of EyePhone.
As unveiled over on MIT Technology Review, the EyePhone software has been developed by a team of researchers at Dartmouth College which allows a modern smartphone to be controlled entirely via the eyes – no poking at the screen required.
Although the technology has obvious applications for those who are sick of having to wipe smudgy fingerprints off their nice smartphone display, it also promises to improve accessibility to smartphone technologies for those who have poor fine motor control, are missing digits, paralyzed, or even missing entire limbs.
The software package – currently developed for use on a Nokia 810 smartphone – uses the front-facing camera to take a constant video stream of the user’s face, which it then analyzes to find the location of the eyes. Picking a particular eye to track, the EyePhone software then attempts to calculate what section of the screen is currently being viewed.
This information is then used to highlight an icon or menu option. If the software has made the wrong choice, the user simply adjusts their gaze in the corresponding direction until the selection they were after is highlighted. Triggering the selection – to launch a program, for example, or make a call to a chosen contact – is as simple as blinking.
Interestingly, the software takes an oblique approach to the task: rather than tracking the angle of the eye – i.e. where it is looking – the EyePhone tracks the relative position of the user’s eye to the screen, meaning that in order to move their selection around the display the user actually moves the handset, not their eyes.
Project leader Professor Andrew Campbell is quick to admit that this is far from the first demonstration of eye-tracking technology as a human-computer interface, but states that “existing algorithms were highly inaccurate in mobile conditions–even if you are standing and there’s a small movement in your arm, you’d get a large amount of blurring and error” which would prevent the software from correctly tracking your eye movement.
Campbell’s team was able to correct for this by including a training process which uses images taken of the user’s eyes in both indoor and outdoor lighting to correct for changing light conditions – and the results are quite convincing, with the team claiming a 76 percent accuracy rage outdoors dropping to a still impressive 60 percent if the user is walking while using the EyePhone software.
It will be a while before the EyePhone technology finds its way into our smartphone handsets, but eye tracking is certainly a technology which is likely to become a hot topic over the next few years.
