DiffCR: A Fast Conditional Diffusion Framework for Cloud Removal from Optical Satellite Images

This repository is the official PyTorch implementation of the TGRS 2024 paper DiffCR.

Xuechao Zou^1*, Graduate Student Member, Kai Li^2*, Student Member, IEEE, Junliang Xing², Senior Member, IEEE,
Yu Zhang¹, Shiying Wang¹, Lei Jin³, Pin Tao^1,2,†, Member, IEEE

Qinghai University¹ , Tsinghua University²,
Beijing University of Posts and Telecommunications³

Requirements

To install dependencies:

pip install -r requirements.txt

To download datasets:

• Sen2_MTC_Old: multipleImage.tar.gz

• Sen2_MTC_New: CTGAN.zip

Training

To train the models in the paper, run these commands:

python run.py -p train -c config/ours_sigmoid.json

Test

To test the pre-trained models in the paper, run these commands:

python run.py -p test -c config/ours_sigmoid.json

Evaluation

To evaluate my models on two datasets, run:

python evaluation/eval.py -s [ground-truth image path] -d [predicted-sample image path]

Citation

@ARTICLE{diffcr,
  author={Zou, Xuechao and Li, Kai and Xing, Junliang and Zhang, Yu and Wang, Shiying and Jin, Lei and Tao, Pin},
  journal={IEEE Transactions on Geoscience and Remote Sensing}, 
  title={DiffCR: A Fast Conditional Diffusion Framework for Cloud Removal From Optical Satellite Images}, 
  year={2024},
  volume={62},
  number={},
  pages={1-14},
}

Results

Quantitative Results

• Ablation Study

• Main Results

Qualitative Results

• Visualization results on the Sen2_MTC_Old dataset.

---

---

---

---

• Visualization results on the Sen2_MTC_New dataset.

---

---

---

---