Research

Paper

TESTING March 25, 2026

UW-VOS: A Large-Scale Dataset for Underwater Video Object Segmentation

Authors

Hongshen Zhao, Jingkang Tai, Yuhang Wu, Wenkang Zhang, Xi Lan, Shangyan Wang, Tianyu Zhang, Wankou Yang

Abstract

Underwater Video Object Segmentation (VOS) is essential for marine exploration, yet open-air methods suffer significant degradation due to color distortion, low contrast, and prevalent camouflage. A primary hurdle is the lack of high-quality training data. To bridge this gap, we introduce $\textbf{UW-VOS}$, the first large-scale underwater VOS benchmark comprising 1,431 video sequences across 409 categories with 309,295 mask annotations, constructed via a semi-automatic data engine with rigorous human verification. We further propose $\textbf{SAM-U}$, a parameter-efficient framework that adapts SAM2 to the underwater domain. By inserting lightweight adapters into the image encoder, SAM-U achieves state-of-the-art performance with only $\sim$2$\%$ trainable parameters. Extensive experiments reveal that existing methods experience an average 13-point $\mathcal{J}\&\mathcal{F}$ drop on UW-VOS, while SAM-U effectively bridges this domain gap. Detailed attribute-based analysis further identifies small targets, camouflage, and exit-re-entry as critical bottlenecks, providing a roadmap for future research in robust underwater perception.

Metadata

arXiv ID: 2603.24006
Provider: ARXIV
Primary Category: cs.CV
Published: 2026-03-25
Fetched: 2026-03-26 06:02

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2603.24006v1</id>\n    <title>UW-VOS: A Large-Scale Dataset for Underwater Video Object Segmentation</title>\n    <updated>2026-03-25T07:10:54Z</updated>\n    <link href='https://arxiv.org/abs/2603.24006v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2603.24006v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>Underwater Video Object Segmentation (VOS) is essential for marine exploration, yet open-air methods suffer significant degradation due to color distortion, low contrast, and prevalent camouflage. A primary hurdle is the lack of high-quality training data. To bridge this gap, we introduce $\\textbf{UW-VOS}$, the first large-scale underwater VOS benchmark comprising 1,431 video sequences across 409 categories with 309,295 mask annotations, constructed via a semi-automatic data engine with rigorous human verification. We further propose $\\textbf{SAM-U}$, a parameter-efficient framework that adapts SAM2 to the underwater domain. By inserting lightweight adapters into the image encoder, SAM-U achieves state-of-the-art performance with only $\\sim$2$\\%$ trainable parameters. Extensive experiments reveal that existing methods experience an average 13-point $\\mathcal{J}\\&amp;\\mathcal{F}$ drop on UW-VOS, while SAM-U effectively bridges this domain gap. Detailed attribute-based analysis further identifies small targets, camouflage, and exit-re-entry as critical bottlenecks, providing a roadmap for future research in robust underwater perception.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.CV'/>\n    <published>2026-03-25T07:10:54Z</published>\n    <arxiv:primary_category term='cs.CV'/>\n    <author>\n      <name>Hongshen Zhao</name>\n    </author>\n    <author>\n      <name>Jingkang Tai</name>\n    </author>\n    <author>\n      <name>Yuhang Wu</name>\n    </author>\n    <author>\n      <name>Wenkang Zhang</name>\n    </author>\n    <author>\n      <name>Xi Lan</name>\n    </author>\n    <author>\n      <name>Shangyan Wang</name>\n    </author>\n    <author>\n      <name>Tianyu Zhang</name>\n    </author>\n    <author>\n      <name>Wankou Yang</name>\n    </author>\n  </entry>"
}