CFNet: Optimizing Remote Sensing 🗺️ Change Detection 🕵 through Content-Aware Enhancement

Official repository for the paper "CFNet: Optimizing Remote Sensing Change Detection through Content-Aware Enhancement".

[🤗Datasets] [🤗Checkpoints]

🐣 News

• [2025.3.11] We release the code and checkpoints for CFNet 🚀
• [2025.3.11] We release the arixv paper 🚀

🤔 Addressing Style Variations in Change Detection

Change detection plays a crucial role in remote sensing, enabling the identification and analysis of temporal changes in the same geographical area. However, bi-temporal remote sensing images often exhibit significant style variations due to differences in acquisition conditions. These unpredictable variations pose a challenge to deep neural networks (DNNs), affecting their ability to accurately detect changes.

To address the problem above, we propose Content Focuser Network (CFNet). CFNet achieves state-of-the-art performance on three well-known change detection datasets: CLCD (F1: 81.41%, IoU: 68.65%), LEVIR-CD (F1: 92.18%, IoU: 85.49%), and SYSU-CD (F1: 82.89%, IoU: 70.78%). 🚀

The main contributions of our work:

• Content-Aware strategy, a novel content-based constraint learning strategy that enhances the model's focus on intrinsic content features while reducing the impact of style variations, thereby improving the accuracy and robustness of bi-temporal change detection in remote sensing imagery.
• Focuser module, a novel mechanism that dynamically reweights features to focus on both changed and unchanged areas, leveraging their mutual constraints to enhance parameter regularization and improve model accuracy.

The visualization results on the CLCD dataset demonstrate the great performance of CFNet. For better readability, we present only the results from CLCD here. For a comprehensive view, including visualizations on all three datasets, please refer to our paper.

To further illustrate the effectiveness of the Content-Aware strategy in extracting content features, we visualize the largest-scale feature maps output by the Content Decoder. Since the LEVIR-CD dataset primarily focuses on building changes, the content features predominantly represent structural information related to buildings).

🏁 Get Start

Installation

1. Clone the repository:

git clone https://github.com/wifiBlack/CFNet.git
cd CFNet

2. Create the environment

conda env create -f requirements.yml
conda activate cfnet_env

3. Prepare the checkpoints

git clone https://huggingface.co/wifibk/CFNet
mv ./CFNet ./checkpoints

Playground

You can debug playground.py step by step to gain a deeper understanding of CFNet. Enjoy exploring it! 🥰

Prepare Datasets

You can download the datasets from CFNet-Datasets.

We provide a preprocessed version of the original datasets to make it easier for you to experiment with CFNet!

You can simply run following commands to get data from huggingface repository. The data will be downloaded under CFNet folder.

# CLCD && CLCD-processed
python get_data.py CLCD.tar.gz
python get_data.py CLCD-processed.tar.gz

# LEVIR-CD && LEVIR-CD-processed
python get_data.py LEVIR_CD.tar.gz
python get_data.py LEVIR-CD-processed.tar.gz

# SYSU-CD
python get_data.py SYSU-CD.tar.gz

Train

Run the following command to train CFNet.

# CLCD dataset
python run.py --data-dir CLCD-processed \
              --log-dir logs_clcd \
              --gpu 0 \
              --epochs 100 \
              --batch-size 32 \
              --num-workers 16 \
              --lr 0.0005 
            #   --checkpoints \

# LEVIR-CD dataset
python run.py --data-dir LEVIR-CD-processed \
              --log-dir logs_levircd \
              --gpu 0 \
              --epochs 100 \
              --batch-size 32 \
              --num-workers 16 \
              --lr 0.0005 
            #   --checkpoints \

# SYSU-CD dataset
python run.py --data-dir SYSU-CD \
              --log-dir logs_sysucd \
              --gpu 0 \
              --epochs 100 \
              --batch-size 32 \
              --num-workers 16 \
              --lr 0.0005 
            #   --checkpoints \

Test

Run the following command to test the performance of CFNet.

# CLCD dataset
python test.py \
 --data-dir CLCD \
 --gpu 0 \
 --batch-size 8 \
 --num-workers 8 \
 --checkpoint checkpoints/clcd.pth

 # LEVIR-CD dataset
python test.py \
 --data-dir LEVIR-CD \
 --gpu 0 \
 --batch-size 8 \
 --num-workers 8 \
 --checkpoint checkpoints/levir-cd.pth

 # SYSU-CD dataset
python test.py \
 --data-dir SYSU-CD \
 --gpu 0 \
 --batch-size 8 \
 --num-workers 8 \
 --checkpoint checkpoints/sysu-cd.pth

✅ Citation

If this project contributes to your research or applications, please consider citing it using the following BibTeX:

@misc{wu2025cfnetoptimizingremotesensing,
      title={CFNet: Optimizing Remote Sensing Change Detection through Content-Aware Enhancement}, 
      author={Fan Wu and Sijun Dong and Xiaoliang Meng},
      year={2025},
      eprint={2503.08505},
      archivePrefix={arXiv},
      primaryClass={cs.CV},
      url={https://arxiv.org/abs/2503.08505}, 
}