Integrating Semi-Supervised and Active Learning for Semantic Segmentation

TL;DR

This paper introduces a combined semi-supervised and active learning framework for semantic segmentation that reduces annotation costs and improves accuracy by selectively refining pseudo-labels and focusing manual labeling on uncertain regions.

Contribution

It presents a novel integrated approach with a pseudo-label auto-refinement module and selective manual labeling, outperforming existing methods on benchmark datasets.

Findings

Outperforms state-of-the-art methods in semantic segmentation benchmarks.

Effectively refines pseudo-labels without increasing labeling budget.

Focuses manual labeling on the most uncertain regions for efficiency.

Abstract

In this paper, we propose a novel active learning approach integrated with an improved semi-supervised learning framework to reduce the cost of manual annotation and enhance model performance. Our proposed approach effectively leverages both the labelled data selected through active learning and the unlabelled data excluded from the selection process. The proposed active learning approach pinpoints areas where the pseudo-labels are likely to be inaccurate. Then, an automatic and efficient pseudo-label auto-refinement (PLAR) module is proposed to correct pixels with potentially erroneous pseudo-labels by comparing their feature representations with those of labelled regions. This approach operates without increasing the labelling budget and is based on the cluster assumption, which states that pixels belonging to the same class should exhibit similar representations in feature space. Furthermore, manual labelling is only applied to the most difficult and uncertain areas in unlabelled data, where insufficient information prevents the PLAR module from making a decision. We evaluated the proposed hybrid semi-supervised active learning framework on two benchmark datasets, one from natural and the other from remote sensing imagery domains. In both cases, it outperformed state-of-the-art methods in the semantic segmentation task.