Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

FARCLUSS: Fuzzy Adaptive Rebalancing and Contrastive Uncertainty Learning for Semi-Supervised Semantic Segmentation

About

Semi-supervised semantic segmentation (SSSS) faces persistent challenges in effectively leveraging unlabeled data, such as ineffective utilization of pseudo-labels, exacerbation of class imbalance biases, and neglect of prediction uncertainty. Current approaches often discard uncertain regions through strict thresholding favouring dominant classes. To address these limitations, we introduce a holistic framework that transforms uncertainty into a learning asset through four principal components: (1) fuzzy pseudo-labeling, which preserves soft class distributions from top-K predictions to enrich supervision; (2) uncertainty-aware dynamic weighting, that modulate pixel-wise contributions via entropy-based reliability scores; (3) adaptive class rebalancing, which dynamically adjust losses to counteract long-tailed class distributions; and (4) lightweight contrastive regularization, that encourage compact and discriminative feature embeddings. Extensive experiments on benchmarks demonstrate that our method outperforms current state-of-the-art approaches, achieving significant improvements in the segmentation of under-represented classes and ambiguous regions.

Ebenezer Tarubinga, Jenifer Kalafatovich, Seong-Whan Lee• 2025

Related benchmarks

TaskDatasetResultRank
Semantic segmentationCityscapes (val)
mIoU81
572
Semantic segmentationPascal VOC 21 classes (val)
mIoU80.3
103
Semantic segmentationPascal Blended 1/4 augmented (train)
mIoU79.8
32
Semantic segmentationPascal Blended augmented (1/8 train)
mIoU79
32
Semantic segmentationPascal Blended 662 labels augmented (1/16 train)
mIoU78.5
31
Semantic segmentationPascal VOC Classic 2012 (1/8)
Unlabeled Samples per Epoch9.2
7
Semantic segmentationPascal VOC Classic 2012 (1/4)
Unlabeled Samples/Epoch7.90e+3
7
Semantic segmentationCityscapes 1/16
Unlabeled Samples/Epoch2.7
7
Semantic segmentationCityscapes (1/8)
Unlabeled Samples per Epoch2.6
7
Semantic segmentationCityscapes 1/4
Unlabeled Samples/Epoch2.20e+3
7
Showing 10 of 13 rows

Other info

Follow for update