TECH NEWS – Facebook’s artificial intelligence research team came up with interesting technology.

Cornell University, which is in Ithaca, New York, United States, has published a research document on its website about a project called Vid2Game, which would be powered by two AI networks. One of them, Pose2Pose, would take input signals from a controller for example, and manipulate a given pose based on it. The other one, Pose2Frame, would use a dynamic background image and generate a high-resolution frame.

„In this work, we develop a novel method for extracting a character from an uncontrolled video sequence and then reanimating it, on any background, according to a 2D control signal. Our method is able to create long sequences of coarsely-controlled poses in an autoregressive manner.

These poses are then converted into a video sequence by a second network, in a way that enables the careful handling and replacement of the background, which is crucial for many applications. Our work paves the way for new types of realistic and personalized games, which can be casually created from everyday videos. In addition, controllable characters extracted from YouTube-like videos can find their place in the virtual worlds and augmented realities,” the document says.

However, seeing something like this from Facebook, who is facing a multi-billion dollar fine in the United States (which possibly doesn’t even hit them hard financially by this point…), isn’t promising at all. Who knows what they could use these videos for in the background? Sure, you could go and make games with them, but it feels a bit like Big Brother already, and let us not forget that Facebook is one of the biggest tech companies of the world at the moment…

Let’s see in the video below how Vid2Game works.

Source: WCCFTech