New Approaches to Image and Video Reconstruction Using Deep Learning at Facebook at F8 2019
We often want to improve our images and videos, such as increasing their resolution or adding color to black & white film. Much progress has been made in recent years through deep learning, and specifically the use of generative adversarial networks (GANs). However, GANs can be slow, and both difficult and expensive to train. In this session, we’ll show you how to colorize old black & white movies and drastically increase the resolution of microscopy images using new PyTorch-based tools from fast.ai, the Salk Institute, and DeOldify that can be trained in just a few hours on a single GPU.