Creating an AR solution that meets the many requirements for performance, ergonomics and cost has been a major challenge for all in the field – let alone finding killer applications that can drive large volumes and cost reductions. A new Israeli-based company called EyeWay is now coming out of stealth to offer their solutions to meet these challenges.
The company will launch a new Silicon-valley based subsidiary headed by Nikhil Balram, who leaves Google where he oversaw all display-related developments. He is also well known to the display community as an expert in AR/VR/MR technology. Having personally worked with Nikhil in the past, I have high regard for his technical knowledge and business acumen. If he is leaving Google to head a small company, I have got to believe they have something important.
The company is not revealing too much about their technology yet, but they will plan to start disclosing more in the coming months. Once it is safe for in-person meetings, I am sure they will begin showing off their technology as well.
In the company’s debut press release, co-founder and CEO of EyeWay Vision Ltd., Rani Cohen, said “EyeWay’s truly immersive AR experience is based on its unique approach of eye tracking, direct retinal projection, and foveated projection which yields super-bright images even outdoors, with an ultra-wide perceived Field of View.” While I don’t have any inside knowledge of their solution, there are a number of keywords here that hint at what they are working on (plus I am sure their patents will disclose more).
For example, each of the items detailed by Cohen have been tried by various developers in AR headsets. But in most cases, optimizing for that feature can cause compromises in other performance, ergonomic or cost elements of the solution. As Balram said in a brief conversation we had,
“You can’t just buy and assemble a bunch of components to create a solution, you need a system-level approach that considers many factors and their trade-offs.”
For example, many AR solutions are focused on using waveguides to deliver the image to the eye. These are very optically inefficient requiring very bright light sources – which can increase cost and heat. Foveated rendering can help reduce the rendering load of the system, but you still often need a very high-resolution display solution (the exception being Varjo with a two-display solution). Direct retinal projection implies a laser-scanned solution which has noted issues with contrast, uniformity and legibility at least in the Hololens 2 product. Wide field-of-view often comes at the expense of pixels per degree, thus reducing the fidelity of the image.
So what do all these hints suggest? It sounds like a laser-scanned system that can optimize the resolution and brightness based on where the user’s eyes are looking and can adjust the field of view as well. On their web site, they also mention “variable focus” suggesting that the vergence- accommodation issue can also be largely mitigated by changing where objects appear, perhaps informed by the eye tracking.
I will be very interested in learning more as the company moves forward with its plans. (CC)
Chris Chinnock is an industry consultant and writer serving clients in emerging display ecosystems such as mini/micro-LED, 8K, light field/holographic, AR/VR/MR and content production. He also serves as executive director of the 8K Association. He can be reached at [email protected].