On Feburary 28th, Facebook revealed a method that more widely available transforms 2D photos into 3D constructs, rolling it out to most iPhone and Android users.
The announcement was made on the Facebook AI blog some moments ago, where the engineers of the organization went into the scope of how exactly they pulled this off.
The tool, called 3D Images, was previously available only for high-end phones with multiple cameras or for desktop users who wanted to build their own depth map to direct the 3D rendering process.
For anyone with an iPhone 7 or higher, or a recent midrange or better Android device, the feature is available now, says Facebook. Obviously Facebook isn’t the first one to use AI to infer 3D data from a 2D image. Google has been doing that for years with the Pixel phones.
You should first upgrade your Facebook app to the latest version, to check it out for yourself. Tap Your What’s on Your Mind after that? Post the prompt of formation at the top of your News Feed and scroll through the options below.
You will see that there is now a 3D picture option available-tap it and it will take you to your camera roll where you can pick an image, select the image you like, and Facebook will convert it into 3D, after which you will receive a simulated 3D posting.
This is just a simulation, and the feature won’t make pictures like those taken by a dual-lens camera fully three-dimensional. But, when you zoom, rotate, or tilt your screen it will still react to seeing it from a few different angles.
Facebook used machine learning to infer a picture’s three-dimensional structure, filling in the depths your single-camera lens was unable to catch on your 2D frame. Previously, this could only be achieved through phones that had two or more cameras, and could take portrait mode shots.
The feature does not appear to have just rolled out for everyone yet, though. It seems to be a server-side update but, just to be sure, you can also upgrade your Facebook app to the latest version.