AR is getting smaller to make a bigger impact

Augmented reality (AR) is an interactive experience that mixes the real world with computer generated perceptual information by overlaying digital images onto real-world views. In doing this, AR gives users an enhanced visual experience. However, as tech ideas get bigger, devices are getting smaller. Most devices, goggles or head-up displays for example, require complex multi-lens systems are too big and no good for use in everyday eyeglasses. So how do you integrate AR into everyday life?

The challenge of size and image quality

Traditional AR systems use multiple lenses to project and refine images which adds bulk and it is not something that is easily reducible in size without hefty compromises in performance. Shrinking a typical four-lens system into something as small as eyeglasses will often reduce the quality of computer-generated images and limit the field of view because, the smaller the lenses, the less they are able to project clear, sharp images.

Researcher led by Youguang Ma have taken on this challenge, and their findings, published in ACS Photonics, propose a hybrid AR display design that combines metasurfaces and refractive lenses with microLED technology, drastically reducing the size of the AR components while maintaining image quality.

How hybrid optics transform AR eyewear

A metasurface is an ultrathin, lightweight silicon nitride film etched with intricate patterns designed to manipulate light. In this case, the metasurface helps focus and shape the light emitted from a microLED display. The refractive lens, made from a synthetic polymer, then sharpens and refines the image by correcting distortions and aberrations. This process creates a more compact, single-lens AR system that can be incorporated into regular eyewear.

The microLED display has tiny arrays of green LEDs that project the initial image, and the metasurface shapes this light into a black-and-green image on the refractive lens, which then refines the image further before projecting it onto a real-world surface or object. The researchers also used a computational algorithm that corrects minor optical imperfections to improve the overall visual output.

Computational image reinforcement: better visuals

The research team developed an algorithm that pre-processes the image to correct any optical distortions caused by the lens system. By identifying and adjusting for these minor imperfections, the algorithm improves the final projected image to make it sharper and more accurate.

Testing the hybrid AR display, the researchers integrated it into a pair of eyeglasses, and the projected images from a one-lens system had less than 2% distortion across a 30° field of view – which is on par with current commercial AR systems that typically use four lenses. Also, the algorithm improved the structural similarity of reprojected test images by 4%, achieving a similarity score of 74.3% between the projected image and the original.

AR in daily life

The current prototype only works with green light, but the researchers are hoping that future versions could incorporate full-colour displays. If successful, this system could make AR technology small enough for everyday glasses meaning a door to combining AR into personal devices could open up. More than that, it could be beneficial in surgery, for example, by using real-time overlays of patient data or surgical guides. In autonomous driving, it could project critical navigation and safety information onto the driver’s field of view without requiring them to look away from the road. While it is still in its early stages, this technology could eventually make AR a part of everyday life​.