A new facial recognition technology is allowing users to make anyone on video say anything in real-time.
The software known as Face2Face was designed by a team of researchers from Max Planck Institute for Informatics, the University of Erlangen-Nuremberg and Stanford University.
Similar to how the popular smartphone app Snapchat offers a feature that allows users to “face swap” with a friend in the same camera frame, the new software allows a source actor to control the mouth and facial expressions of any person on video without compromising their physical identity.
To showcase their project, the team published a video of their new technology in action. In the video, they show an actor speaking to a webcam while his facial expressions and speech patterns are mimicked in real-time by targets on YouTube videos, like Donald Trump, Vladimir Putin, Barack Obama and George Bush.
Check out the video below to see how the real-time facial reenactment software works…
The project is set to be presented at the Washington Convention Center in this year’s IEEE Conference on Computer Vision and Pattern Recognition. But right now, the new technology is just in the research stage and is not yet publicly available, according to Mashable.
Learn more about Face2Face from the team behind the project here.