We already can’t believe photos because of Photoshop.
Now, we can’t believe videos either, especially videos of political figures, because of technological “advancements” that enable alterations of their facial features and expressions in real time — the actual time during which the video is screened.
Imagine someone manipulating a video of you to make you appear to be speaking, frowning, smiling, laughing or angry, when you aren’t doing any of that.
ZeroHedge reports, April 9, 2016, that a recently published paper, titled “Face2Face: Real-time Face Capture and Reenactment of RGB Videos,” by a team of researchers (Justus Thies, Michael Zollhöffer, Marc Stamminger, Christian Theobalt, and Matthias Nießner) from the University of Erlangen-Nuremberg, Max Planck Institute for Informatics, and Stanford University, shows how disturbingly easy it is for a “surrogate actor,” in real time, to “reenact” and change the facial expressions of someone in a video, by using simple tools of a commodity webcam and some software.
From the paper’s abstract, in typical academic jargon:
We present a novel approach for real-time facial reenactment of a monocular target video sequence (e.g., Youtube video). The source sequence is also a monocular video stream, captured live with a commodity webcam. Our goal is to animate the facial expressions of the target video by a source actor and re-render the manipulated output video in a photo-realistic fashion. To this end, we first address the under-constrained problem of facial identity recovery from monocular video by non-rigid model-based bundling. At run time, we track facial expressions of both source and target video using a dense photometric consistency measure. Reenactment is then achieved by fast and efficient deformation transfer between source and target. The mouth interior that best matches the re-targeted expression is retrieved from the target sequence and warped to produce an accurate fit. Finally, we convincingly re-render the synthesized target face on top of the corresponding video stream such that it seamlessly blends with the real-world illumination. We demonstrate our method in a live setup, where Youtube videos are reenacted in real time.
In simple English, what the paper says is that they can seamlessly and convincingly fake anyone on TV talking, reacting, answering questions, and emoting — in real time.
Here’s the team’s video, demonstrating their real-time “face capture and reenactment” of Donald Trump, George W. Bush, and Vladimir Putin.
The researchers claim that their purpose is “purely research-focused” — “to demonstrate the capabilities of modern computer vision and graphics technology, and convey it in an approachable and fun way.”
They point out that “computer-generated videos” have been used in entertainment movies for over 30 years. “Virtually every high-end movie production contains a significant percentage of synthetically-generated content (from Lord of the Rings to Benjamin Button),” the results of which “are hard to distinguish from reality” and “often goes unnoticed that the content is not real.” What makes the researchers’ work different from Hollywood’s is that “we can edit pre-recorded videos in real-time on a commodity PC.”
While the purpose of the creators of this new technology may be that of pure science, now that the technology is here, it certainly will be used by people who are less scrupulous. The conclusion is that we can no longer believe what we see or hear in videos. Just imagine what opponents can and will do to videos of Donald Trump or other 2016 presidential candidates.