Eventually the need for special low background steel will be just a memory. Oddly enough, steel isn’t the only thing that we’ve dragged from the bottom of the ocean.

However, Google might just have a close-enough approximation before long. The Google Brain team has developed a system that uses neural networks to fill in the details on very low-resolution images.

In conjunction with Princeton University School of Architecture’s 44 Low-resolution Houses exhibition, curated by Michael Meredith and designed by MOS, this multi-day international symposium expands ...

The system learns by taking in thousands of pairs of images – one at full resolution, the other downsampled to a jagged, low-res image. It pores over these pairs to work out which filters it can ...