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

TL;DR

FARCLUSS introduces a comprehensive semi-supervised segmentation framework that leverages uncertainty and class imbalance to improve segmentation accuracy, especially for under-represented and ambiguous regions.

Contribution

It proposes a novel combination of fuzzy pseudo-labeling, uncertainty-aware weighting, adaptive rebalancing, and contrastive regularization for semi-supervised segmentation.

Findings

Outperforms state-of-the-art methods on benchmark datasets.

Improves segmentation of under-represented classes.

Effectively utilizes uncertain regions rather than discarding them.

Abstract

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.