Learning Joint Denoising, Demosaicing, and Compression from the Raw Natural Image Noise Dataset

TL;DR

This paper presents RawNIND, a diverse raw image dataset, and introduces two denoising methods that outperform traditional approaches, demonstrating the benefits of raw data workflows for image denoising, compression, and generalization.

Contribution

The paper introduces RawNIND dataset and proposes two novel raw data denoising methods that improve performance and efficiency over traditional developed image approaches.

Findings

RawNIND enables better generalization across sensors and workflows.

Raw data denoising improves rate-distortion performance.

Integrated raw denoising and compression enhances efficiency.

Abstract

This paper introduces the Raw Natural Image Noise Dataset (RawNIND), a diverse collection of paired raw images designed to support the development of denoising models that generalize across sensors, image development workflows, and styles. Two denoising methods are proposed: one operates directly on raw Bayer data, leveraging computational efficiency, while the other processes linear RGB images for improved generalization to different sensors, with both preserving flexibility for subsequent development. Both methods outperform traditional approaches which rely on developed images. Additionally, the integration of denoising and compression at the raw data level significantly enhances rate-distortion performance and computational efficiency. These findings suggest a paradigm shift toward raw data workflows for efficient and flexible image processing.