Home

Awesome

Restorative Step-Calibrated Diffusion

This is the code repository for our paper Step-Calibrated Diffusion for Biomedical Optical Image Restoration:

@misc{lyu2024stepcalibrated,
      title={Step-Calibrated Diffusion for Biomedical Optical Image Restoration}, 
      author={Yiwei Lyu and Sung Jik Cha and Cheng Jiang and Asadur Chowdury and Xinhai Hou and Edward Harake and Akhil Kondepudi and Christian Freudiger and Honglak Lee and Todd C. Hollon},
      year={2024},
      eprint={2403.13680},
      archivePrefix={arXiv}
}

The full paper and full appendix are available in full_paper_and_appendix.pdf. Please note that due to the size of the file, Github preview may not display all figures properly. Please download the pdf and view locally.

To train the model, use the following command:

python3 main.py --config configs/config.yaml

To fine-tune the model, use the following command:

python3 main.py --config configs/config-finetune.yaml

To train the classifier, run

python3 classifier.py

To run inference using RSCD, run the following command:

python3 inference/inference_recalibrate.py --config configs/config-inference-recalibrate.py