SFIR

The official implementation code for the paper "SFIR: Optimizing Spatial and Frequency Domains for Image Restoration"

1. Training the Model

To train the model, you can use the following command:

python main.py --mode train --data_dir path/xxx

Replace path/xxx with the path to your training dataset directory. This command will start the model training process and train the model using the provided dataset.

2. Testing the Model

To test a pre-trained model, use the following command:

python main.py --mode test --data_dir path/xxx --test_model path_to_ckpt

• path/xxx: The directory path to your test dataset.
• path_to_ckpt: The path to the checkpoint file of the pre-trained model.

This command will load the specified model and evaluate its performance on the test dataset.

3. Results

The complete code and results are being compiled, so stay tuned.

4. Acknowledgments

Our code is based on the IRNeXt architecture. We want to express our sincere thanks to the authors for their outstanding work.

If this code repository is helpful to you, please cite our article.

@article{gu2025sfir,
  title={SFIR: Optimizing Spatial and Frequency Domains for Image Restoration},
  author={Gu, Yubin and Meng, Yuan and Chen, Siting and Ji, Jiayi and Sun, Xiaoshuai and Ruan, Weijian and Ji, Rongrong},
  journal={Pattern Recognition},
  pages={112188},
  year={2025},
  publisher={Elsevier}
}