Sometimes looking at a photo or watching a video as it was filmed doesn’t really provide the viewer with context, like where was the photo or video taken? Was it inside the house? What does the house look like? However it seems that Facebook has created an AI that can create a 3D world just based on photos and videos.

This was revealed during Facebook’s F8 event where Rachel Franklin, Facebook’s head of social VR demonstrated the tech on stage. Basically it takes information from photos and videos, and based on that, it will attempt to construct a 3D world that could potentially be where the photo or video was taken.

The end result? While the “world” itself seems to be made up of splotches, it does seem to be pretty spot on. Is it accurate? We suppose at a glance it does look like it did a good job of roughly reconstructing the room. This is based on tech that Facebook is calling “photogrammetry” which the company claims is a good way to explore your memories.

For example imagine looking at childhood photos while being in a virtual reconstruction of the home you grew up in, or your favorite restaurant, and so on. That being said, Facebook claims that this is a prototype and are keeping mum on their plans on whether or not they will implement it in the future. However in a statement made to Engadget, “We think in the future, people may be able to use technology like this to revisit moments and places that are important to them, with people they care about.”

Filed in General. Read more about .

Discover more from Ubergizmo

Subscribe now to keep reading and get access to the full archive.

Continue reading