Watch: These scientists made it incredibly easy for anyone to doctor JNU videos
New technology means you can literally put words in someone else's mouth.
Doctored videos have never really been a major concern in India, at least not until recently. As anyone who tries to watch a Bollywood movie with computer generated imagery can tell you, it takes a huge budget to make something that is fake look real. Even the ones that rocked the national media in February used incredibly primitive editing techniques to make it look like Jawaharlal Nehru University students were shouting allegedly sedition slogans.
All that is about to change.
A research collaboration between scientists at Stanford University, the Max Planck Institute and the University of Erlangen-Nuremberg in Germany have created a software that allows them to control the face of just about anyone in any video. All the program needs is about 15 seconds video of someone's face, after which it creates a 3D model that it can easily manipulate.
As the professors show in the video, the program can take the face of someone, like former US President George W Bush or JNU Student Union President Kanhaiya Kumar. It then maps out their face through the 3D modeling. After that, a test subject sitting in a lab can use his or her own face to manipulate that 3D model and add their own voice to it.
Effectively, technology that would have once cost millions and taken a huge computer generated imagery lab to put together for a big budget movie is now possible with a program and just 15 seconds of video. The possibilities, nefarious or otherwise, are endless.
