USIS16K: High-Quality Dataset for Underwater Salient Instance Segmentation

2025-06-24

AI Summary

Underwater salient instance segmentation (USIS) has long been hindered by challenging imaging conditions and severe scarcity of high-quality, large-scale annotated data. To address this, we introduce USIS16K—the first large-scale, high-resolution benchmark dataset for USIS—comprising 16,151 diverse underwater images with pixel-accurate instance masks across 158 object categories. The dataset spans multiple scenes, illumination conditions, and object poses, significantly enhancing data diversity and task scalability. Annotations are generated via rigorous human verification and multi-source acquisition, enabling joint evaluation of object detection and salient instance segmentation. We further release a suite of open-source baseline models, which achieve state-of-the-art performance across multiple underwater vision tasks. USIS16K is publicly available, establishing a critical foundation for advancing underwater visual understanding, model development, and standardized evaluation.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Large Multimodal Models (LMMs)
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Data management and stream processing for Web, mobile and wireless applications
• Search and Retrieval-Augmented AI: Web search models and ranking
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Inspired by the biological visual system that selectively allocates attention to efficiently identify salient objects or regions, underwater salient instance segmentation (USIS) aims to jointly address the problems of where to look (saliency prediction) and what is there (instance segmentation) in underwater scenarios. However, USIS remains an underexplored challenge due to the inaccessibility and dynamic nature of underwater environments, as well as the scarcity of large-scale, high-quality annotated datasets. In this paper, we introduce USIS16K, a large-scale dataset comprising 16,151 high-resolution underwater images collected from diverse environmental settings and covering 158 categories of underwater objects. Each image is annotated with high-quality instance-level salient object masks, representing a significant advance in terms of diversity, complexity, and scalability. Furthermore, we provide benchmark evaluations on underwater object detection and USIS tasks using USIS16K. To facilitate future research in this domain, the dataset and benchmark models are publicly available.

Problem

• Lack of large-scale dataset for underwater salient instance segmentation
• Challenges in underwater environments limit USIS research
• Need for diverse, high-quality annotated underwater object images

Innovation

• Large-scale dataset USIS16K with 16,151 images
• High-quality instance-level salient object masks
• Benchmark evaluations for underwater object tasks