Blog

Artificial Intelligence can now sharpen blurry pictures up to 60 times

Artificial intelligence now has the capacity to unveil the person behind blurred or pixelated images, making it more difficult to disguise a person’s identity.

Software engineers have used machine learning to teach a piece of software to adapt image recognition techniques to recognise objects, faces and words in obscured images. Now, researchers from Duke University have announced that they have a tool that can turn blurry images of people’s faces into perfect computer-generated portraits in high definition.

There have been methods prior to this that can scale an image of a face up to eight times its original resolution, but the Duke team has developed a more refined AI tool that can create realistic-looking face with up to 64 times the resolution.

“Never have super-resolution images been created at this resolution before with this much detail,” said Duke computer scientist Cynthia Rudin, who led the team.

Instead of taking a low-resolution image and slowly adding new detail, the system scours AI-generated examples of high-resolution faces, searching for ones that look as much as possible like the input image when shrunk down to the same size, essentially creating a new face that don’t exist.

See the results and upload images for yourself at http://pulse.cs.duke.edu/.