arXiv cs.AI· Quang-Trung Truong, Thanh-Nghia Nguyen, Phu-Phong Vo, Jaeyeon Kim, Cheng Chen, Jie Zhang, Binh-Son Hua, Duc Thanh Nguyen, Ivor Tsang, Sai-Kit Yeung·· 13 小时前AI 评分20
MarineLife-16K：面向运动感知海洋视频理解的高属性视频-文本数据集
MarineLife-16K: An Attribute-Rich Video-Text Dataset for Motion-Aware Marine Video Understanding
AI 导读
MarineLife-16K 是一个大规模视频-文本数据集，包含 2,000 个视频-文本对，标注 13 类属性，覆盖每场景最多四个对象、对象视觉/几何属性、行为、环境上下文与相机运动，横跨 7 个国家及多种光照与时段，并由此生成 16,080 个视频-问题对（12,080 道多选、4,000 道开放题）。
正文 · 原文
Abstract:Marine video understanding is crucial for a wide range of real-world applications. However, existing marine video datasets suffer from limitations, including the limited number of data samples, fine-grained text descriptions, and detailed visual attributes. To address these limitations, we introduce a large-scale, video-text dataset featuring rich, in-the-wild visual attributes tailored for complex marine environments. Our dataset comprises 2,000 video-text pairs annotated with 13 attributes covering up to four objects per scene, object visual/geometric properties, behaviors, environmental context, and camera motion, spanning diverse lighting conditions, times of day, and 7 countries. From these annotations, we generate 16,080 video-question pairs (12,080 multiple-choice, 4,000 open-ended). Furthermore, we benchmark both proprietary and open-source video-LLM models on video question answering and video captioning tasks. We find that (i) proprietary models often achieves superior performances; (ii) VLMs handle object motion well but struggle with complex camera motion; (iii) closed-source models struggle to perceive camera motion; and (iv) VLMs struggle to reason in static cameras.
| Comments: | Published at WACV2027 |
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.33478 [cs.CV] |
| (or arXiv:2610.33478v1 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2610.33478 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Quang Trung Truong [view email]
[v1]
Sat, 24 Oct 2026 06:05:26 UTC (21,091 KB)
来源：arXiv cs.AI · arxiv.org