Rethinking Image Compression on the Web with Generative AI

TL;DR

This paper introduces a generative AI-based image compression framework that significantly reduces bandwidth usage while maintaining perceptual image quality, offering a promising alternative to traditional methods for web images.

Contribution

It presents a novel AI-driven image reconstruction approach using text prompts and conditioning inputs, achieving high compression ratios with minimal perceptual quality loss.

Findings

Achieves up to 99.8% bandwidth savings in optimal cases

Maintains high perceptual similarity compared to traditional compression

User study confirms effective preservation of image meaning and structure

Abstract

The rapid growth of the Internet, driven by social media, web browsing, and video streaming, has made images central to the Web experience, resulting in significant data transfer and increased webpage sizes. Traditional image compression methods, while reducing bandwidth, often degrade image quality. This paper explores a novel approach using generative AI to reconstruct images at the edge or client-side. We develop a framework that leverages text prompts and provides additional conditioning inputs like Canny edges and color palettes to a text-to-image model, achieving up to 99.8% bandwidth savings in the best cases and 92.6% on average, while maintaining high perceptual similarity. Empirical analysis and a user study show that our method preserves image meaning and structure more effectively than traditional compression methods, offering a promising solution for reducing bandwidth usage and improving Internet affordability with minimal degradation in image quality.