High-Fidelity Generative Image Compression
This work addresses the challenge of high-fidelity image compression for applications requiring perceptual quality and efficiency, representing a significant but incremental improvement in the field.
The paper tackles the problem of generative lossy image compression by combining Generative Adversarial Networks with learned compression, achieving state-of-the-art results where their method is preferred over previous approaches even when using less than half the bitrate.
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-distortion-perception theory and practice by evaluating our approach both quantitatively with various perceptual metrics, and with a user study. The study shows that our method is preferred to previous approaches even if they use more than 2x the bitrate.