How Facebook & Oculus Redefined the Future of Social VR at F8
New technologies showcased at F8 could prove to be the boost that social VR needs to enter the mainstream.
The word ‘presence’ saw liberal use with regards to virtual reality (VR) at the F8 conference. Facebook, and by extension Oculus, are moving towards a goal of a form of social VR that allows two or more people to feel like they are occupying the same space, despite potentially being thousands of miles apart.
Facebook has been a bastion of social media for over a decade now, allowing people all over the world to connect and communicate. Not surprising that the company wants to take this further by utilising new technologies.
As Mike Schroepfer said at the Day 2 keynote of the F8 conference, “We think VR is the only way we’ll get people to connect over very large distances.” as he introduced some of the tools, apps and services which are aimed at this specific goal.
Oculus TV, Rooms and Venues
for many people one of the most valuable and precious aspects of friendship is to just be hanging out in the same room. Whether this is watching TV or films together, playing videogames, listening to music or just chatting, these moments build bonds. Oculus are trying to provide this experience in VR with Oculus Rooms and Oculus TV, which allows people to virtually share a space, chat and watch TV together. Oculus Venues lets you take your friends to a virtual concert, allowing friends and relatives who might be far apart to share in the experience together.
Much attention was given to the improvements that Facebook had made to the avatars used in Facebook Spaces, a social app which has received a great deal of attention. When the app was first unveiled, avatars consisted for a featureless blue head and disembodied hands.
In the current version, users can take a photo of themselves and see it transformed into a reasonably accurate avatar using a process that relies on AI.
This is not the end of the road, however. One of the more challenging aspects of avatar ‘presence’ involves lip sync. As any fan of animation or foreign films will tell you, getting the lip sync wrong can produce a significant uncanny valley effect. To combat this, Facebook has again turned to AI to help process speech and provide lip sync, not just in English but in multiple languages.
As Schroepfer commented during his keynote: ‘Hands are such an important part of how we express emotions to each other, but they are surprisingly difficult to capture.’
Facebook have built some proof-of-concept systems that allow for hands to be properly and accurately captured, but that is only the first step towards digitising a user’s entire body for VR for the best possible experience.
One of the impressive demonstrations shown at F8 was a 3D image capture of a room, which had been recreated in CGI. The recreated room was so convincing that much of the audience had difficulty telling apart the live-action video from the 3D recreation.
A 3D depth map was used to accurately capture the various dimensions and textures of each object. Even the mirrors in the room were able to reflect the room as they would in real life.
If this level of realism is able to be integrated into the next level of VR software, it could well be a game changer for how VR is regarded.