"Now, artificial intelligence gobbles up substantial computational resources (and battery life) every time you glance at your phone to unlock it with face ID. In the future, one piece of glass could recognize your face without using any power at all.
“This is completely different from the typical route to machine vision,” says [Zongfu] Yu.
He envisions pieces of glass that look like translucent squares. Tiny strategically placed bubbles and impurities embedded within the glass would bend light in specific ways to differentiate among different images. That’s the artificial intelligence in action.
For their proof of concept, the engineers devised a method to make glass pieces that identified handwritten numbers. Light emanating from an image of a number enters at one end of the glass, and then focuses to one of nine specific spots on the other side, each corresponding to individual digits.
The glass was dynamic enough to detect, in real-time, when a handwritten 3 was altered to become an 8.
Designing the glass to recognize numbers was similar to a machine-learning training process, except that the engineers “trained” an analog material instead of digital codes. Specifically, the engineers placed air bubbles of different sizes and shapes as well as small pieces of light-absorbing materials like graphene at specific locations inside the glass.
“We could potentially use the glass as a biometric lock, tuned to recognize only one person’s face” says Yu. “Once built, it would last forever without needing power or internet, meaning it could keep something safe for you even after thousands of years.”
My only concern is that the high optical power needed for non-linear operations in "NN glass" might burn the person's face. However, if recognition is achieved in nanosecond time with a short laser pulse, this might not be an issue:
Interesting and different! Zongfu is a smart guy (and former collaborator). Hope this finds a good application.
ReplyDeleteI was actually impressed by the "smart glass" response in the video when a handwritten "3" was converted to "8". Very good demonstration of how NN works.
ReplyDeleteWell, the demo is nice, but question remains: can it be practically implemented? Probably not. Definitely not at the thin glass like shown in the video.
ReplyDeleteIt is quite interesting. It seems to be a typical learning cycle. As the algorithm is first implemented in digital to mimic the analog in the nature. Then a more customized analog way is invented to increase the efficiency.
ReplyDeleteIf they can find a cheap way to manufacture it, it could be a fair good solution for some fixed tasks.