Duke University researchers have developed an AI tool that can turn blurry, unrecognizable pictures of people’s faces into high resolution pictures, in finer detail than ever before.
old methods can scale an image of a face up to x8 times to its original resolution. But the Duke team has come up with a way to take a handful of pixels and create realistic-looking faces with up to 64 times the resolution
“Never have super-resolution images been created at this resolution before with this much detail,”Duke computer scientist Cynthia Rudin, Team Leader
Their system cannot be used to identify people, the researchers say: It won’t turn an out-of-focus, unrecognizable photo from a security camera into a crystal clear image of a real person. Rather, it is capable of generating new faces that don’t exist, but look plausibly real.
Their system can convert a 16×16-pixel image of a face to 1024 x 1024 pixels in a few seconds, The researchers will present their method, called PULSE, at the 2020 Conference on Computer Vision and Pattern Recognition (CVPR), held remotely from June 14 to June 19.
Check there work from here http://pulse.cs.duke.edu/.