Remember when Facebook CEO Mark Zuckerberg tried to show off his company’s clunky Spaces VR tech by taking a virtual tour of hurricane-hit Puerto Rico in 2017? This is nothing like that.
The company has quietly been developing a revolutionary system to create virtual avatars of people that closely resemble them in their facial features and mannerisms. It’s not ready for prime time, but as Wired noted, it’s already surprisingly sophisticated.
The company is calling them Codec Avatars, and works on the technology behind them at its Facebook Reality Labs (FRL, formerly Oculus Research) facility in Pittsburgh. Here’s an example of how well this system works; in the clip below, two people wearing VR headsets and having a conversation are on the left of the frame, while their avatars are on the right:
To pull this off, the team at FRL uses a contraption dubbed Mugsy. It’s a dome fitted with 132 camera lenses and 350 lights, and they’re all aimed at the center, where a subject must be seated before their facial features can be mapped over the course of an hour. A second area with even more cameras captures how your body moves, in order to help animate your avatar even more realistically.
The <3 of EU tech
The latest rumblings from the EU tech scene, a story from our wise ol' founder Boris, and some questionable AI art. It's free, every week, in your inbox. Sign up now!
We don’t yet know when, or if, we’ll see a consumer-facing product come out of this Facebook project, but it’s a major milestone for modeling human behavior in virtual environments. You can find out more about this technology in Wired’s article, where Peter Rubin delves into its implications, and shares his apprehension of how it might be misused in this new age of misinformation.
TNW Conference 2019 is coming! Check out our glorious new location, inspiring line-up of speakers and activities, and how to be a part of this annual tech extravaganza by clicking here.
Get the TNW newsletter
Get the most important tech news in your inbox each week.