sRGB Real Noise Modeling
via Noise-Aware Sampling with Normalizing Flows

Dongjin Kim*, Donggoo Jung*, Sungyong Baik, Tae Hyun Kim
*Equal Contribution

[ICLR2024] Paper

Camera / ISO	Clean	Noisy	Sampled Noisy
Samsung Galaxy S6 Edge / 03200			0032-0023.mp4
Google Pixel / 06400			0018-0017.mp4

Table of Contents

• Framework Overview
• How to run
  • Installation
  • Training and Evaluation
• Results
• Citation
• Acknowledgement

Framework Overview

Training

Inference

How to run

Installation

# Clone this repo
git clone https://github.com/dongjinkim9/NAFlow.git
cd NAFlow

# Create and activate conda environment
conda env create -f environments.yaml
conda activate naflow

Training and Evaluation

Model	Training Instructions	Testing Instructions
NAFlow	Link	Link
Denoising Networks	-	Link

Results

NAFlow

Denoising networks with NAFlow

Citation

If you find our work useful in your research, please consider citing our paper:

@article{kim2023srgb,
  title={sRGB Real Noise Modeling via Noise-Aware Sampling with Normalizing Flows},
  author={Kim, Dongjin and Jung, Donggoo and Baik, Sungyong and Kim, Tae Hyun},
  booktitle={ICLR},
  year={2023}
}

Acknowledgement

The codes are based on DeFlow. We thank the authors for sharing their codes.