How to Exploit the Transferability of Learned Image Compression to Conventional Codecs

TL;DR

This paper explores leveraging learned image compression and generative adversarial networks to optimize images for conventional codecs, achieving over 20% rate improvements without additional decoding overhead.

Contribution

It introduces a method to use learned filters and GANs to modify images for better compression efficiency with existing codecs, without increasing decoding complexity.

Findings

Achieves over 20% rate reduction for MS-SSIM distortion.

Remodels conventional codecs to improve compression without extra decoding cost.

Performs favorably on task-aware image compression tasks.

Abstract

Lossy image compression is often limited by the simplicity of the chosen loss measure. Recent research suggests that generative adversarial networks have the ability to overcome this limitation and serve as a multi-modal loss, especially for textures. Together with learned image compression, these two techniques can be used to great effect when relaxing the commonly employed tight measures of distortion. However, convolutional neural network based algorithms have a large computational footprint. Ideally, an existing conventional codec should stay in place, which would ensure faster adoption and adhering to a balanced computational envelope. As a possible avenue to this goal, in this work, we propose and investigate how learned image coding can be used as a surrogate to optimize an image for encoding. The image is altered by a learned filter to optimise for a different performance measure or a particular task. Extending this idea with a generative adversarial network, we show how entire textures are replaced by ones that are less costly to encode but preserve sense of detail. Our approach can remodel a conventional codec to adjust for the MS-SSIM distortion with over 20% rate improvement without any decoding overhead. On task-aware image compression, we perform favourably against a similar but codec-specific approach.