Smart Glasses Recognize Images without Any Sensors

The sophis­ticated tech­nology that powers face recognition in many modern smartphones someday could receive a high-tech upgrade that sounds and looks sur­prisingly low-tech. This window to the future is none other than a piece of glass. Uni­versity of Wisconsin-Madison engineers have devised a method to create pieces of smart glass that can recognize images without requiring any sensors or circuits or power sources. “We’re using optics to condense the normal setup of cameras, sensors and deep neural networks into a single piece of thin glass,” says Zongfu Yu.

Tiny impurities in smart glass mimic artificial intelligence by bending light to recognize different numbers. (Source: Z. Yu)

Embedding arti­ficial intel­ligence inside inert objects is a concept that, at first glance, seems like something out of science fiction. However, it’s an advance that could open new frontiers for low-power elec­tronics. Now, arti­ficial intel­ligence gobbles up substantial compu­tational resources 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 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.

“The fact that we were able to get this complex behavior with such a simple structure was really something,” says Erfan Khoram, a graduate student in Yu’s lab. 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’re accustomed to digital computing, but this has broadened our view,” says Yu. “The wave dynamics of light propa­gation provide a new way to perform analog arti­ficial neural computing“.

One such advantage is that the compu­tation is completely passive and intrinsic to the material, meaning one piece of image-recog­nition glass could be used hundreds of thousands of times. “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.”

Additionally, it works at literally the speed of light, because the glass distin­guishes among different images by distorting light waves. Although the up-front training process could be time consuming and compu­tationally demanding, the glass itself is easy and inex­pensive to fabricate. In the future, the researchers plan to determine if their approach works for more complex tasks, such as facial recognition.

“The true power of this tech­nology lies in its ability to handle much more complex classification tasks instantly without any energy consumption,” says Ming Yuan, a colla­borator from Columbia University. “These tasks are the key to create arti­ficial intelli­gence: to teach driverless cars to recognize a traffic signal, to enable voice control in consumer devices, among numerous other examples.”

Unlike human vision, which is mind-bogglingly general in its capa­bilities to discern an untold number of different objects, the smart glass could excel in specific appli­cations – for example, one piece for number recog­nition, a different piece for identifying letters, another for faces, and so on. “We’re always thinking about how we provide vision for machines in the future, and imagining appli­cation specific, mission-driven tech­nologies.” says Yu. “This changes almost everything about how we design machine vision.” (Source: U. Wisconsin-Madison)

Reference: E. Khoram et al.: Nanophotonic media for artificial neural inference, Phot. Res. 7, 823 (2019); DOI: 10.1364/PRJ.7.000823

Link: Photonics Lab, University of Wisconsin Madison, Madison, USA

Speak Your Mind