# Neural Image Decompression: Learning to Render Better Image Previews

**Authors:** Shumeet Baluja, Dave Marwood, Nick Johnston, Michele Covell

arXiv: 1812.02831 · 2018-12-10

## TL;DR

This paper introduces a neural decoding method that enhances the quality of extremely compressed image previews without changing the encoding, improving both quantitative metrics and semantic content preservation.

## Contribution

It presents a neural-based decoding approach that significantly improves image reconstruction quality at ultra-low bitrates while maintaining compatibility with existing encoding standards.

## Key findings

- Higher PSNR and SSIM scores than traditional methods
- Better preservation of semantic content in reconstructed images
- Compatible with existing image encoding streams

## Abstract

A rapidly increasing portion of Internet traffic is dominated by requests from mobile devices with limited- and metered-bandwidth constraints. To satisfy these requests, it has become standard practice for websites to transmit small and extremely compressed image previews as part of the initial page-load process. Recent work, based on an adaptive triangulation of the target image, has shown the ability to generate thumbnails of full images at extreme compression rates: 200 bytes or less with impressive gains (in terms of PSNR and SSIM) over both JPEG and WebP standards. However, qualitative assessments and preservation of semantic content can be less favorable. We present a novel method to significantly improve the reconstruction quality of the original image with no changes to the encoded information. Our neural-based decoding not only achieves higher PSNR and SSIM scores than the original methods, but also yields a substantial increase in semantic-level content preservation. In addition, by keeping the same encoding stream, our solution is completely inter-operable with the original decoder. The end result is suitable for a range of small-device deployments, as it involves only a single forward-pass through a small, scalable network.

## Full text

_Full body text omitted from this summary view._ Fetch the complete paper as Markdown: https://tomesphere.com/paper/1812.02831/full.md

## Figures

39 figures with captions in the complete paper: https://tomesphere.com/paper/1812.02831/full.md

## References

39 references — full list in the complete paper: https://tomesphere.com/paper/1812.02831/full.md

---
Source: https://tomesphere.com/paper/1812.02831