CPDM: Achieving Superior Accuracy in Semi-Supervised Remote Sensing Segmentation

conda create -n CPDM python=3.8
pip install -r requirements.txt

Refer to the preparation of the following code repository

https://github.com/xiaoqiang-lu/WSCL.git

├── ./pretrained
    └── resnet101.pth
    
├── [RsDataSets/Vaihingen(DFC22/GID15/...)]
    ├── images
    └── labels

It is recommended to use Visual Studio Code as the compiler.

cd SemiExp/DATASETNAME
python train.py

We thank WSCL, LSST, and UniMatch, for part of their codes, processed datasets, data partitions, and pretrained models.