Google is stepping up efforts to develop a mixed reality headset in the vein of Meta and Apple’s projects, The Verge reports on January 20. Baptized “Iris Project”, its primary use would be augmented reality, but it would do so by a method called “passthrough”, that is to say with the help of integrated cameras that transcribe the environment on the screens of a virtual reality headset.
Stadia’s streaming technology put to good use
This approach is simpler to implement and provides better experiences than projecting images on semi-transparent glasses, as Microsoft HoloLens, Magic Leap or Snap do with its Spectacles 4. Informants from The Verge indicate that the prototype being developed resembles a ski mask and is made to work on its own, without being connected to another device or an external battery.
On the other hand, it should rely on the cloud to do “remote rendering”, that is to say calculate heavy scenes in a data center and stream them to the headset via a cellular connection. Google can rely on its cloud gaming technology, which is already at the heart of Stadia, to do this. The advantage of this system is that it sharply reduces the power required on the device and draws less on its battery.
No launch until at least 2024
The headset will use a System-on-a-Chip customized by Google, like the latest smartphone in its Pixel range, and is currently running Android. It should be noted, however, that the company has just hired Mark Lucovsky from Meta to develop an operating system dedicated to augmented reality.
The device would still be at the beginning of its development phase and its commercialization – if it takes place – therefore remains distant. The Verge’s sources point to a planned release in 2024, but which could very well be delayed (as is the case for Apple). The team in charge of its development has about 300 people so far, but Google plans to recruit hundreds more. At its head is Clay Bavor, who has been leading Google’s AR/VR efforts for several years now.
In particular, he was at the helm of the Daydream platform, with which Google had tried to counter Meta’s efforts in virtual reality in 2016. A half-effort that had not convinced. He is now in charge of the “Labs” department, which develops Iris, but also Starline, the Area 120 incubator, and a blockchain project.
Starline still in development
About Starline precisely, Google would apparently like to market its system at the same time as Iris, or sometime in 2024 if everything goes as it hopes. As a reminder, this is an immersive communication system coupling a battery of cameras and microphones with a lightfield display system to make immersive video calls without having to wear a headset. Since the technology is expensive (tens of thousands of dollars for each installation) and cumbersome to set up, it is currently reserved for a few large companies.
Google bears the scars of past failures
These new projects, Starline and especially Iris, mark the return of Google in the race for virtual and augmented reality. The company became interested in the subject very early: as early as 2012 with Google Glass for augmented reality, which was a failure for several reasons, then the Tango project, and finally Cardboard and Daydream for VR, which did not keep their promises either. She had also invested very early in Magic Leap. After these setbacks, the company had focused on the software part, especially with Lens.
It should be noted that Google also owns North, a Canadian start-up that was developing lightweight augmented reality glasses but with limited functionality. A concept closer to that of Google Glass, but better executed. It is likely that this project will continue quietly in addition to the rest.
This return to the field of hardware in AR / VR is done in a certain way by force of circumstances. Meta and Apple are investing heavily in it, and Microsoft is also making no secret of its ambitions. Google could hardly afford to completely ignore this technology, even if it has burned its wings on it in the past. These repeated failures – and his propensity not to invest completely in a project, and then cancel it without persevering – nevertheless complicate his task. It is now lagging behind the competition and it may well fail again if it does not get serious about it, especially since its credibility with third-party developers is being undermined.