Generating Training Data for Denoising Real RGB Images via Camera Pipeline Simulation

TL;DR

This paper presents a camera pipeline simulation method to generate realistic RGB image noise for training denoising models, significantly improving their performance on real-world images compared to traditional AWGN models.

Contribution

The authors develop a novel simulation pipeline that accurately models camera noise and degradation, enabling more effective training of denoising neural networks for real RGB images.

Findings

Realistic noise modeling improves denoising performance by 3 dB.

Simulating denoising and demosaicking is crucial for realistic training.

The proposed pipeline outperforms traditional AWGN-based training methods.

Abstract

Image reconstruction techniques such as denoising often need to be applied to the RGB output of cameras and cellphones. Unfortunately, the commonly used additive white noise (AWGN) models do not accurately reproduce the noise and the degradation encountered on these inputs. This is particularly important for learning-based techniques, because the mismatch between training and real world data will hurt their generalization. This paper aims to accurately simulate the degradation and noise transformation performed by camera pipelines. This allows us to generate realistic degradation in RGB images that can be used to train machine learning models. We use our simulation to study the importance of noise modeling for learning-based denoising. Our study shows that a realistic noise model is required for learning to denoise real JPEG images. A neural network trained on realistic noise outperforms the one trained with AWGN by 3 dB. An ablation study of our pipeline shows that simulating denoising and demosaicking is important to this improvement and that realistic demosaicking algorithms, which have been rarely considered, is needed. We believe this simulation will also be useful for other image reconstruction tasks, and we will distribute our code publicly.