Home

Awesome

RealSRT

Paper | BibTex

PyTorch implementation of ECCV 2024 paper "Confidence-Based Iterative Generation for Real-World Image Super-Resolution"

Introduction

<div align=center> <img src="./figures/teaser.png" width="70%" height="70%">

Visualizations of our confidence-based iterative generation process for real-world SR.

</div>

Method

<div align=center> <img src="./figures/pipeline.png">

Overview of RealSRT.

</div>

Installation

This implementation is based on BasicSR

git clone https://github.com/USTC-JialunPeng/RealSRT
cd RealSRT
pip install -r requirements.txt
python setup.py develop

Inference

  1. Download the pre-trained model and place it in ./experiments/pretrained_models/

  2. Download the test dataset (e.g., RealSR), place input images in /data/input/ and place target images (if available) in /data/target/

  3. Testing

python inference_realsrt.py --input /data/input/ --output /data/results/ --model_path experiments/pretrained_models/net_g_80000.pth
  1. To reproduce scores in Table 1, run
python calculate_metrics.py

Citing

If our method is useful for your research, please consider citing.

@inproceedings{peng2024confidence,
  title={Confidence-Based Iterative Generation for Real-World Image Super-Resolution},
  author={Peng, Jialun and Luo, Xin and Fu, Jingjing and Liu, Dong},
  booktitle={European Conference on Computer Vision},
  year={2024}
}