yzygit1230 / hatnet

HATNet

• The pytorch implementation for HATNet in paper "Hybrid Attention-aware Transformer Network Collaborative Multiscale Feature Alignment for Building Change Detection".

Requirements

• Python 3.6
• Pytorch 1.7.0

Datasets Preparation

The path list in the datasest folder is as follows:

|—train

• ||—A

• ||—B

• ||—OUT

|—val

• ||—A

• ||—B

• ||—OUT

|—test

• ||—A

• ||—B

• ||—OUT

where A contains pre-temporal images, B contains post-temporal images, and OUT contains ground truth images.

Train

• python train.py --dataset-dir dataset-path

Test

• python eval.py --ckp-paths weight-path --dataset-dir dataset-path

Visualization

• python visualization visualization.py --ckp-paths weight-path --dataset-dir dataset-path (Note that batch-size must be 1 when using visualization.py)
• Besides, you can adjust the parameter of full_to_color to change the color

Citation

If this work is helpful to you, please cite it as:

@article{xu2024hybrid,
  title={Hybrid Attention-Aware Transformer Network Collaborative Multiscale Feature Alignment for Building Change Detection},
  author={Xu, Chuan and Ye, Zhaoyi and Mei, Liye and Yu, Haonan and Liu, Jianchen and Yalikun, Yaxiaer and Jin, Shuangtong and Liu, Sheng and Yang, Wei and Lei, Cheng},
  journal={IEEE Transactions on Instrumentation and Measurement},
  volume={73},
  pages={1--14},
  year={2024},
  publisher={IEEE}
}