High-Fidelity Generative Image Compression
PubDate: Dec, 2020
Teams: ETH Zürich, Google Research
Writers: Fabian Mentzer, George Toderici, Michael Tschannen, Eirikur Agustsson
We extensively study how to combine Generative Adversarial Networks and learned compression to obtain a state-of-the-art generative lossy compression system. In particular, we investigate normalization layers, generator and discriminator architectures, training strategies, as well as perceptual losses. In contrast to previous work, i) we obtain visually pleasing reconstructions that are perceptually similar to the input, ii) we operate in a broad range of bitrates, and iii) our approach can be applied to high-resolution images. We bridge the gap between rate-distortionperception theory and practice by evaluating our approach both quantitatively withvarious perceptual metrics, and with a user study. The study shows that our method is preferred to previous approaches even if they use more than 2× the bitrate.