Artificial intelligence system can turn low-resolution images into HD versions
AI has done many wonders in the recent past. Take Google Assistant for example, a very powerful AI packed inside your Android firmware. In the recent studies, scientists have used AI for multiple purposes. Combining AI with Machine Learning would give the most reliable results. Recently, an AI system was used to turn low-resolution images into HD versions of the original ones.
Many attempts were made to increase the resolution of the images in the past. Such a technique was named as Single-Image-Super-Resolution technology (SISR). Traditionally, they used to add few extra pixels by taking the average of those existing ones. But the results were poor and pictures turned blurry.
With AI anything can be made possible and scientists at Max Planck Institute of Intelligent Systems, Germany finally did it. This AI tech can predict the appropriate pixels and add them to the picture to make it look rich and crisp.
Researchers at Max Planck University say, that they have trained the algorithm with thousands of images and the results they obtained with this AI powered tech (EnhanceNet-PAT technology) is remarkable.
In most of the smartphones produced today, they come with something called Low-Light photography. This technique is very similar to the one used in the SISR technologies. People are getting more involved in experiencing better image quality. Be it a smartphone or a DSLR. This up-sampling technology proposed by researchers at Max Planck University is definitely going to make a huge change in how we see the images.