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

Ebenezer Tarubinga · Jenifer Kalafatovich · Seong-Whan Lee

Pattern Recognition & Machine Learning Lab, Korea University

IJCNN 2025

Overview of the CW-BASS framework. Stage I generates confidence-weighted pseudo-labels with dynamic thresholding. Stage II refines segmentation with boundary-aware learning and confidence decay.

65.9%
mIoU · Cityscapes 1/30
78.4%
mIoU · Cityscapes 1/4
76.2%
mIoU · Pascal VOC 1/4
100
Labeled Images (3.3%)

Abstract

Semi-supervised semantic segmentation (SSSS) aims to improve segmentation performance by utilizing large amounts of unlabeled data with limited labeled samples. Existing methods often suffer from coupling, where over-reliance on initial labeled data leads to suboptimal learning; confirmation bias, where incorrect predictions reinforce themselves repeatedly; and boundary blur caused by limited boundary-awareness and ambiguous edge cues. To address these issues, we propose CW-BASS, a novel framework for SSSS. We assign confidence weights to pseudo-labels to mitigate the impact of incorrect predictions, and leverage boundary-delineation techniques that remain underutilized in SSSS. Extensive experiments on Pascal VOC 2012 and Cityscapes demonstrate that CW-BASS achieves state-of-the-art performance. Notably, CW-BASS achieves a 65.9% mIoU on Cityscapes under a challenging 1/30 split (100 images), highlighting its effectiveness in limited-label settings.

Key Contributions

Confidence-Weighted Loss

Reduces coupling by adjusting pseudo-label influence based on predicted confidence scores, downweighting unreliable predictions.

Dynamic Thresholding

Mitigates confirmation bias with an adaptive mechanism that learns to filter pseudo-labels based on model performance.

Boundary-Aware Module

Tackles boundary blur using explicit boundary supervision to refine segmentation near object edges.

Confidence Decay

Reduces label noise through a progressive strategy that refines pseudo-label confidence during training.

Visual Results (Pascal VOC 2012)

Sample 1

Sample 2

Sample 3

Sample 4

Sample 5

Sample 6

Quantitative Results

Performance comparison on Pascal VOC 2012 and Cityscapes

Performance comparison (mIoU %) on Pascal VOC 2012 and Cityscapes across different label partitions. All experiments use DeepLabV3+ with ResNet-50. Best results in bold.

Qualitative Results

Qualitative comparison of segmentation outputs

Qualitative comparison on Pascal VOC 2012 and Cityscapes. CW-BASS produces sharper boundaries and more accurate segmentation masks compared to prior methods.

Citation

@inproceedings{tarubinga2025cwbass,
  author    = {Tarubinga, Ebenezer and Kalafatovich, Jenifer
               and Lee, Seong-Whan},
  title     = {CW-BASS: Confidence-Weighted Boundary-Aware Learning
               for Semi-Supervised Semantic Segmentation},
  booktitle = {2025 International Joint Conference on Neural
               Networks (IJCNN)},
  year      = {2025},
  pages     = {1--8},
  address   = {Rome, Italy},
  doi       = {10.1109/IJCNN64981.2025.11227871},
  keywords  = {Semi-Supervised Learning, Semantic Segmentation,
               Pseudo-Labeling, Confidence Weighting}
}