The faces above don’t seem remarkable. They could easily come from Facebook or LinkedIn. In fact, they were dreamt up by a new kind of AI algorithm, Will Knight writes.
Fake faces: Nvidia researchers created these faces using a new sort of generative adversarial network (more on GANs here). The images released show how much of an improvement the researchers’ new work is.
A big quality leap: The team were inspired by a technique known as style transfer to build their GAN in a different way, letting their algorithm identify different elements of a face, which the researchers could then control, remixing elements like age, race, gender and even freckles.
Uses: GANs are likely to change the way video games and special effects are generated and can be used to sharpen up degraded images or video. But the work also shows how advances in machine learning could lead to all sorts of new possibilities for fakery and misinformation. Read the full story here.
See the full story here: https://www.technologyreview.com/s/612612/these-incredibly-real-fake-faces-show-how-algorithms-can-now-mess-with-us/