CW-BASS: Confidence-Weighted Boundary Aware Learning for Semi-Supervised Semantic Segmentation

This is the official PyTorch implementation of our IJCNN 2025 paper:

CW-BASS: Confidence-Weighted Boundary Aware Learning for Semi-Supervised Semantic Segmentation

Project Page: https://psychofict.github.io/CW-BASS/

Getting Started

Data Preparation

Pre-trained Model

ResNet-50 | DeepLabv2-ResNet-101

Dataset

Pascal JPEGImages | Pascal SegmentationClass | Cityscapes leftImg8bit | Cityscapes gtFine

File Organization

├── ./pretrained
    ├── resnet50.pth
    └── deeplabv2_resnet101_coco_pretrained.pth
    
├── [Your Pascal Path]
    ├── JPEGImages
    └── SegmentationClass
    
├── [Your Cityscapes Path]
    ├── leftImg8bit
    └── gtFine

Training and Testing

export semi_setting='pascal/1_8/split_0'

CUDA_VISIBLE_DEVICES=0,1 python -W ignore main.py \
  --dataset pascal --data-root [Your Pascal Path] \
  --batch-size 16 --backbone resnet50 --model deeplabv3plus \
  --labeled-id-path dataset/splits/$semi_setting/labeled.txt \
  --unlabeled-id-path dataset/splits/$semi_setting/unlabeled.txt \
  --pseudo-mask-path outdir/pseudo_masks/$semi_setting \
  --save-path outdir/models/$semi_setting

Acknowledgement

The image partitions are borrowed from Context-Aware-Consistency and PseudoSeg. Part of the training hyper-parameters and network structures are adapted from ST++ and PyTorch-Encoding.

• ST++: https://github.com/LiheYoung/ST-PlusPlus.
• Context-Aware-Consistency: https://github.com/dvlab-research/Context-Aware-Consistency.
• PyTorch-Encoding: https://github.com/zhanghang1989/PyTorch-Encoding.
• OpenSelfSup: https://github.com/open-mmlab/OpenSelfSup.

Citation

If you find this project useful, please consider citing:

@article{tarubinga2025cw,
  title={CW-BASS: Confidence-Weighted Boundary-Aware Learning for Semi-Supervised Semantic Segmentation},
  author={Tarubinga, Ebenezer and Kalafatovich, Jenifer and Lee, Seong-Whan},
  journal={arXiv preprint arXiv:2502.15152},
  year={2025}
}