Add us to yourGoogle favoritesResearchers have developed an artificial intelligence-based algorithm that significantly enhances the quality of pixelated images. Presented at the International Conference on Computer Vision 2017, and available in a pre-publication version, these findings show that the researchers used machine learning (deep learning). The algorithm, named EnhanceNet-PAT, can generate a high-resolution version of a low-resolution image. Initially, the researchers tried to replicate the original image by precisely reconstructing the pixels, but the results were not satisfactory. The database feeding the deep learning was compiled with numerous images to enable the AI to replicate all types of surfaces.