Slashdot Mirror
eyeBlog
cottonbuds writes "Researchers at Human Media Lab, Queen's University in Canada presented the ECSGlasses: eye contact sensing glasses that report when people look at their wearer. When eye contact is detected, the glasses stream this information to appliances to inform these about the wearer's engagement. According to HML.Blog the ECSGlasses uses a wearable, wireless Eye-Contact Sensor (1.3MB .jpg) to gauge when the user receives eye-contact from an onlooker. eyeBlog uses this information to record and publish face-2-face conversations without dividing the user's attention between the event being recorded, and the device being used to record it. Moreover, because eyeBlog uses eye-contact to start and stop recording, users do not need to sift through hours of footage to find interesting segments. If you are the academic type you can read the paper (2.2MB .pdf), otherwise the video in .mpg (1:49min, 320x240, 7.5MB), or mp4 (1:49min, 320x240, 4.9MB) should explain everything. Video Mirror: .mp4 .mpg."
If you poke around the images on the site you'll see that they have already implemented what you're speaking about. TV, and phones being devices that they have shown being controlled by attention/visual focus.
"If you're flammable and have legs, you are never blocking a fire exit." - Mitch Hedberg
The Context Aware Computing group at the MIT Media Lab produced Eye aRe years ago. These devices detect attention paid to another such unit (yes, everyone needs to be equipped with one, which is a difference from the canadian item), and not just staring in the general direction: There is a demo, for instance, wherein the wearer can look at a computer, which prompts it to unpause movie playback. Looking away--without turning of the head, eyes only--repauses the movie (this is insanely hard to demo without sound; you can't really tell if it has paused or not when you're looking away, so you look back at it. oops.). The Eye-Are units are certainly smaller, at the very least. To be fair, they're using different technologies, but the optical advanced-ness of the canadian unit seems wasted on a supid application. The usefulness of having inanimate objects--say, appliances--know where you're looking, when, and how (the Eye aRe detects blinking, an increase of which can signal any number of things) seems to dwarf that of some hyper-blogging solution. Both devices, of course, offer a sort of unconscious appliance-control possiblilties, but one is much smaller, and cheaper to manufacture (namely, the Eye aRe).