AI turns pixelated photos in crystal-clear images
21 November 2017 | Written by La redazione
Mehdi S. M. Sajjadi, Bernhard Schölkopf e Michael Hirsc could take a grossly pixelated, low-resolution photo and turn it into a crystal-clear image. That’s possible thanks to EnhanceNet classify by its inventors as a “Single Image Super-Resolution Through Automated Texture Synthesis.” That’s exactly what it does: It scales up very low-resolution photos by synthesizing textures that introduce new detail into the resulting high-resolution image. The synthetic detail makes the resulting pictures so realistic that they are almost indistinguishable from the actual photos. The software creates these images by “feed-forward fully convolutional neural networks in an adversarial training setting.” In other words, like other adversarial neural networks, one system generates results while the other one evaluates the accuracy of the result.