Dual-Granularity Semantic Prompting for Language Guidance Infrared Small Target Detection

2025-11-24

AI Summary

Infrared small target detection suffers from weak feature representation and severe background clutter, while existing text-guided methods rely heavily on manual annotations and suffer from inaccurate textual descriptions. To address these issues, this paper proposes DGSPNet, an end-to-end language-prompt-driven network. Its core contributions are: (1) a novel dual-granularity semantic prompting mechanism that jointly leverages coarse-grained textual priors and fine-grained vision–language alignment, enabling annotation-free text guidance; and (2) text-guided channel attention (TGCA) and text-guided spatial attention (TGSA) modules that collaboratively enhance target sensitivity in both low-level texture and high-level semantic features. Extensive experiments demonstrate that DGSPNet achieves state-of-the-art performance on three benchmark datasets, significantly improving detection accuracy and robustness under complex background conditions.

Technology Category

• Natural Language Processing: Sentence-level Semantics, Textual Inference, etc.
• Computer Vision: Language and Vision
• Machine Learning: Multimodal Learning

Application Category

• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• Web Mining and Content Analysis: Normalization, clustering, classification, and summarization of Web text
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

Abstract

Infrared small target detection remains challenging due to limited feature representation and severe background interference, resulting in sub-optimal performance. While recent CLIP-inspired methods attempt to leverage textual guidance for detection, they are hindered by inaccurate text descriptions and reliance on manual annotations. To overcome these limitations, we propose DGSPNet, an end-to-end language prompt-driven framework. Our approach integrates dual-granularity semantic prompts: coarse-grained textual priors (e.g., 'infrared image', 'small target') and fine-grained personalized semantic descriptions derived through visual-to-textual mapping within the image space. This design not only facilitates learning fine-grained semantic information but also can inherently leverage language prompts during inference without relying on any annotation requirements. By fully leveraging the precision and conciseness of text descriptions, we further introduce a text-guide channel attention (TGCA) mechanism and text-guide spatial attention (TGSA) mechanism that enhances the model's sensitivity to potential targets across both low- and high-level feature spaces. Extensive experiments demonstrate that our method significantly improves detection accuracy and achieves state-of-the-art performance on three benchmark datasets.

Problem

• Improving infrared small target detection accuracy
• Overcoming inaccurate text descriptions in detection methods
• Reducing reliance on manual annotations for target detection

Innovation

• Dual-granularity semantic prompts for language guidance
• Text-guide channel and spatial attention mechanisms
• Visual-to-textual mapping for fine-grained semantic descriptions