Scientists have come up with a solution to convert blurry snaps into high-resolution pictures

A demonstration of how the technology will work on photos (Courtesy: Max Planck Institute) 

The new software promises to upscale small size images into ones that look almost like high-resolution ones.

Researchers at the German Institute have further developed this kind of technology, labelled as SISR, or single-image super-resolution, to give images a realistic texture when magnified from small to large- through the help of Machine Learning.

So what's the gamechanger?

As per the institute, it is artificial intelligence algorithm which learns from its experience as it goes on sharpening it simultaneously for scaling up the quality of the image.

Instead of present image-enhancing technologies based on pixel-perfect reconstruction, it instead aims for faithful texture synthesis, which makes the difference.

“The algorithm is given the task of upsampling millions of low-resolution images to a high-resolution version and is then shown the original, the 'this-is-how-it-should-be'-image. Notice the difference? OK, then learn from your mistake,” explains Mehdi S.M. Sajjadi, one of the co-developers of the technology called EnhanceNet-PAT, the AI driven technology that makes it possible.