Carnegie Mellon researchers create the most convincing deepfakes yet

Ever heard of “deepfakes”? Videos generated with artificial intelligence (AI) that learn to superimpose the face of one person onto the body of another have been used to swap Harrison Ford for Nicolas Cage in countless movie clips, and for far more nefarious purposes, like fake celebrity porn and propaganda. Now, for better or worse, researchers at Carnegie Mellon University have developed a new AI system that’s more powerful — and versatile — than previous attempts.

It’s called “Recycle-GAN,” and the team described it as an “unsupervised, data-driven approach” for transferring the content of one video or photo to another. “Such a content translation and style preservation task has numerous applications, including human motion and face translation from one person to other, teaching robots from human demonstration,” the researchers wrote, “or converting black-and-white videos to color.”

Continue reading… “Carnegie Mellon researchers create the most convincing deepfakes yet”