Lossless Image Compression through Super-Resolution

TL;DR

This paper presents SReC, a lossless image compression method that combines low-resolution storage with iterative super-resolution and entropy coding, achieving state-of-the-art compression rates efficiently.

Contribution

It introduces a novel lossless compression algorithm using super-resolution conditioned on low-res images, improving compression efficiency over existing methods.

Findings

Achieves state-of-the-art compression rates.

Operates efficiently on large datasets.

Utilizes super-resolution conditioned probability prediction.

Abstract

We introduce a simple and efficient lossless image compression algorithm. We store a low resolution version of an image as raw pixels, followed by several iterations of lossless super-resolution. For lossless super-resolution, we predict the probability of a high-resolution image, conditioned on the low-resolution input, and use entropy coding to compress this super-resolution operator. Super-Resolution based Compression (SReC) is able to achieve state-of-the-art compression rates with practical runtimes on large datasets. Code is available online at https://github.com/caoscott/SReC.