This repository is for the paper: 'Semi-supervised Metallographic Image Segmentation via Consistency Regularization and Contrastive Learning'.
- numpy==1.21.5
- opencv_python==4.6.0.66
- tensorboardX==2.6.1
- torch==1.11.0
- torchmetrics==0.10.3
- torchvision==0.12.0
- tqdm==4.64.0
All experiments in our paper were conducted on a single NVIDIA Tesla V100 GPU.
- Install
pip install -r requirements.txt
- Train the model
cd ./scripts/con2net
# on MetalDAM dataset
bash MetalDAM.sh
# on UHCS dataset
bash UHCS.sh
- Test the model
cd ./scripts/test
bash MetalDAM.sh
bash UHCS.sh