Scientists at the Max-Planck Institute for Intelligent Systems in Germany have come up with a clever solution that is capable of producing better results than anything we've seen so far.

The team has developed a tool called EnhanceNet-PAT, which is capable of creating high-definition versions of low-resolution images, using artificial intelligence.

To explain it short - this guys trained big neural network using big amount of very similar images. After this they feed almost same image to neural network and asked it to guess missed pixels in higher resolution. OF course they also hand selected such images and rejected all where network made horrible results.

This approach is not scalable and not really useful for anything outside very specific field. It also frequently can produce unpredictably bad results.

https://arxiv.org/abs/1612.07919