Image Compression, Green-Screens, and Deep Learning
Image compression can get awfully tricky once you get down to really low bit-rates — think 0.1 bpp (bits per pixel) or less. The reigning champ is currently BPG (•), and even it breaks down at really low bpp. This is where Deep Learning comes in, with a bunch of work done on doing adaptive compression by generating textural detail , suppressing artifacts , amongst others. These all work really quite well — 1.7x to 2.5x better than BPG/WebP/JPEG — so well in fact that Machine Learning seems to be the new , and hot!, frontier in image compression these days. The issue though is that most (all?) of them tend to focus on fixing the artifacts that emerge from compression in one form or the other (not that that is a bad thing mind you!). In a new paper, Augustsson et al. (••) have an intriguing new approach using Generative Adversarial Networks (GANs), where they focus on preserving the important parts of the image, and generating the “unimport...