SparseUWSeg: Active Sparse Point-Label Augmentation for Underwater Semantic Segmentation

2025-10-11

AI Summary

Underwater semantic segmentation faces a critical bottleneck: dense pixel-level annotation is prohibitively expensive, while sparse point annotations—though easier to acquire—are hindered by sampling bias and insufficient label propagation. To address this, we propose an active sparse point annotation enhancement framework. First, we design an uncertainty-driven active sampling strategy to improve the discriminative power of selected points. Second, we integrate the SAM2 foundation model with a superpixel-guided hybrid label propagation mechanism to generate fine-grained, structure-preserving pseudo-labels. Third, we incorporate a lightweight interactive annotation tool to enable efficient iterative labeling in ecological underwater scenarios. Evaluated on two public underwater datasets, our method achieves up to a 5.0% mIoU improvement over D+NN under only a 5% point annotation budget, significantly outperforming existing sparse-supervision approaches. The framework thus achieves a favorable trade-off between segmentation accuracy and annotation efficiency.

Technology Category

• Machine Learning: Active Learning
• Computer Vision: Segmentation
• Reasoning under Uncertainty: Uncertainty Representations

Application Category

• Semantics and Knowledge: Representation, semantic annotation, enhancement, enrichments, access and/or integration of a variety of data on the Web
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Semantic segmentation is essential to automate underwater imagery analysis with ecology monitoring purposes. Unfortunately, fine grained underwater scene analysis is still an open problem even for top performing segmentation models. The high cost of obtaining dense, expert-annotated, segmentation labels hinders the supervision of models in this domain. While sparse point-labels are easier to obtain, they introduce challenges regarding which points to annotate and how to propagate the sparse information. We present SparseUWSeg, a novel framework that addresses both issues. SparseUWSeg employs an active sampling strategy to guide annotators, maximizing the value of their point labels. Then, it propagates these sparse labels with a hybrid approach leverages both the best of SAM2 and superpixel-based methods. Experiments on two diverse underwater datasets demonstrate the benefits of SparseUWSeg over state-of-the-art approaches, achieving up to +5% mIoU over D+NN. Our main contribution is the design and release of a simple but effective interactive annotation tool, integrating our algorithms. It enables ecology researchers to leverage foundation models and computer vision to efficiently generate high-quality segmentation masks to process their data.

Problem

• Active sampling strategy maximizes annotation value for underwater segmentation
• Hybrid approach propagates sparse labels using SAM2 and superpixels
• Reduces dependency on expensive dense expert annotations for underwater imagery

Innovation

• Active sampling strategy guides annotation process
• Hybrid approach combines SAM2 and superpixel methods
• Interactive tool integrates algorithms for mask generation