Open-CD is an open source change detection toolbox based on a series of open source general vision task tools.
- 11/17/2022 - Open-CD is upgraded to v0.0.2, requiring a higher version of the MMSegmentation dependency.
- 9/28/2022 - The code, pre-trained models and logs of ChangerEx are available. 😋
- 9/20/2022 - Our paper Changer: Feature Interaction is What You Need for Change Detection is available!
- 7/30/2022 - Open-CD is publicly available!
Supported toolboxes:
Supported change detection model: (The code of some models are borrowed directly from their official repositories.)
- FC-EF (ICIP'2018)
- FC-Siam-diff (ICIP'2018)
- FC-Siam-conc (ICIP'2018)
- STANet (RS'2020)
- IFN (ISPRS'2020)
- SNUNet (GRSL'2021)
- BiT (TGRS'2021)
- ChangeFormer (IGARSS'22)
- TinyCD (arXiv'2022)
- Changer (arXiv'2022)
- ...
Supported datasets:
Please refer to get_started.md in mmseg.
git clone https://github.com/likyoo/open-cd.git
cd open-cd
pip install -v -e .
train
python tools/train.py configs/changer/changer_ex_r18_512x512_40k_levircd.py --work-dir ./changer_r18_levir_workdir --gpu-id 0 --seed 307
infer
# get .png results
python tools/test.py configs/changer/changer_ex_r18_512x512_40k_levircd.py changer_r18_levir_workdir/latest.pth --format-only --eval-options "imgfile_prefix=tmp_infer"
# get metrics
python tools/test.py configs/changer/changer_ex_r18_512x512_40k_levircd.py changer_r18_levir_workdir/latest.pth --eval mFscore mIoU
If you find this project useful in your research, please consider cite:
@article{fang2022changer,
title={Changer: Feature Interaction is What You Need for Change Detection},
author={Sheng Fang and Kaiyu Li and Zhe Li},
year={2022},
eprint={2209.08290},
archivePrefix={arXiv},
primaryClass={cs.CV}
}
@misc{opencd2022,
title={{Open-CD}: An open source change detection toolbox},
author={Open-CD Contributors},
howpublished = {\url{https://github.com/likyoo/open-cd}},
year={2022}
}
Open-CD is released under the Apache 2.0 license.
The inspiration for this project comes from a casual conversation with friends during my internship at Sensetime.