Raw-JPEG Adapter: Efficient Raw Image Compression with JPEG

Mahmoud Afifi, Ran Zhang, Michael S. Brown
9/23/2025
cs.CV

No official code found. Best implementation: SamsungLabs/raw-jpeg-adapter (5 stars, last commit 2025-12-09).

Abstract

Digital cameras digitize scene light into linear raw representations, which the image signal processor (ISP) converts into display-ready outputs. While raw data preserves full sensor information--valuable for editing and vision tasks--formats such as Digital Negative (DNG) require large storage, making them impractical in constrained scenarios. In contrast, JPEG is a widely supported format, offering high compression efficiency and broad compatibility, but it is not well-suited for raw storage. This paper presents RawJPEG Adapter, a lightweight, learnable, and invertible preprocessing pipeline that adapts raw images for standard JPEG compression. Our method applies spatial and optional frequency-domain transforms, with compact parameters stored in the JPEG comment field, enabling accurate raw reconstruction. Experiments across multiple datasets show that our method achieves higher fidelity than direct JPEG storage, supports other codecs, and provides a favorable trade-off between compression ratio and reconstruction accuracy.

View on arXivView PDF

Code Implementations(1)

SamsungLabs/raw-jpeg-adapterCommunity67%

Reference code for the paper: "Raw-JPEG Adapter: Efficient Raw Image Compression with JPEG." Learnable, invertible raw-to-JPEG pre-processing with parameters stored in the JPEG COM (<64 KB), enabling accurate raw reconstruction with much smaller files.

57PythonOct 28, 202510 months agoNOASSERTION

Similar Papers

SnapPix: Efficient-Coding--Inspired In-Sensor Compression for Edge Vision

80% match1 repo22025

Raw2Event: Converting Raw Frame Camera into Event Camera

80% match2025

Catch Missing Details: Image Reconstruction with Frequency Augmented Variational Autoencoder

79% match1 repo472026

LeanVAE: An Ultra-Efficient Reconstruction VAE for Video Diffusion Models

79% match1 repo1162025

Image Compression for Machine and Human Vision with Spatial-Frequency Adaptation

79% match1 repo562026

SCINeRF: Neural Radiance Fields from a Snapshot Compressive Image

79% match1 repo442026

Cite this paper

@article{afifi2025rawjpeg,
  title  = {Raw-JPEG Adapter: Efficient Raw Image Compression with JPEG},
  author = {Mahmoud Afifi and Ran Zhang and Michael S. Brown},
  year   = {2025},
  eprint = {2509.19624},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2509.19624v3}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2509.19624.svg)](https://codeofpaper.com/paper/2509.19624)

Discussion