Eye tracking can be pretty useful for disabled people, but also for non-disabled, as it can be used for interfacing machinery whithouth resorting to hands or voice. The difficult part lies in the tracking of the eyes. There are lots of solutions on the market, the best of which use removable or permanent (in case of monkeys) solenoids attached to the eye. Eye position can then be inferred from the voltage occuring in a 3D magnetic field. Permanent solenoids are bad for MRI, though. Lesser trackers use some sort of optical feedback, sometimes using reflections of light on different ocular surfaces, sometimes using a video of the eyes and some algorithm seraching for the pupils. Those are rather unreliable compared to the solenoid-based ones, though. There is an optical system that is near perfet, but it uses a contact lens with a tiny mirror, an is therefore not permanent. I propose the surgical attachment of a little ring-reflector on the sclera surrounding the iris. Tracking this reflector is easy, and the pattern of reflection can also code where the fovea lies in relation to the ring, making calibration of the eyetracker obsolete.-- loonquawl, May 05 2009 [+] ..but the thought of voluntary eye implants...<shudder>
Oh, and if withouth isn't a word, it should be.-- 2 fries shy of a happy meal, May 05 2009 This seems doable to me without much fuss. One could use a surgery similar to cornea replacement. However, if the technology improves you might be permanently wedded to an old iteration and so something like contacts might be better.-- bungston, May 05 2009 What was wrong with the contact lenses? It seems perfect to me.-- Suede, May 05 2009 I was hoping this would be about measuring the soul via one's optical organs ...-- Aristotle, May 06 2009 [Suede]: The contact lenses are not the soft ones, but the hard ones, and they become uncomfortable rather fast, in fact they are usually applied with a drop of local anesthetic.-- loonquawl, May 06 2009 This is wrong, because it asks humans to adapt to their technology rather than vice versa. A human can tell, rather accurately, where another human is looking: just ask someone to look at a point on your face, and then touch the point where you think they're looking. Software ought to be able to do the same.-- MaxwellBuchanan, May 06 2009 Technology is already beyond that point. Humans can detect whether someone is looking at their eyes or at their ears, for focal points far away from themselves they are not that good. Video-based eye trackers are already better at both - the problem lies in the abolute exactness required (is the person looking at this icon, or the one half a degree besides it?) and the problem of finding the eyes in complex, dynamic environments, an image-recognition task far beyond current capabilities.
Interacting with technology always requires some kind of adaption - carrying a cellphone precludes dancing in the rain, having outlets in the walls precludes sticking a fork anywhere you like, and trying to get from A to B in your personal transport requires you to wrap your kid in a protective shell on the back seat. Implanting something to help technology help you is not unknown (deep stimulation for Parkinson's) as well as implanting something for no practical reason at all (subcutaneous ridges) - so why not implanting something to make life easier? - It is not impairing any natural function, as fun contact lenses do (admittetly to a small degree, but they impair vision).-- loonquawl, May 07 2009 "This is wrong, because it asks humans to adapt to their technology rather than vice versa"
(marked-for-tagline)-- normzone, May 07 2009 random, halfbakery