Semantic-Aware Ship Detection with Vision-Language Integration

TL;DR

This paper introduces a novel semantic-aware ship detection framework that integrates vision-language models with a multi-scale sliding window approach, leveraging a new dataset to improve detection accuracy in complex remote sensing scenarios.

Contribution

The paper presents a new framework combining vision-language models with multi-scale strategies and introduces the ShipSem-VL dataset for fine-grained ship attribute detection.

Findings

Enhanced detection accuracy in complex scenes

Effective semantic attribute recognition

Comprehensive evaluation across multiple tasks

Abstract

Ship detection in remote sensing imagery is a critical task with wide-ranging applications, such as maritime activity monitoring, shipping logistics, and environmental studies. However, existing methods often struggle to capture fine-grained semantic information, limiting their effectiveness in complex scenarios. To address these challenges, we propose a novel detection framework that combines Vision-Language Models (VLMs) with a multi-scale adaptive sliding window strategy. To facilitate Semantic-Aware Ship Detection (SASD), we introduce ShipSem-VL, a specialized Vision-Language dataset designed to capture fine-grained ship attributes. We evaluate our framework through three well-defined tasks, providing a comprehensive analysis of its performance and demonstrating its effectiveness in advancing SASD from multiple perspectives.