Enhancing Infrared Vision: Progressive Prompt Fusion Network and Benchmark

2025-10-10

AI Summary

Existing thermal infrared image enhancement methods typically address isolated degradations—such as noise, low contrast, or blur—in isolation, failing to jointly model their coupled interactions; meanwhile, RGB-based general-purpose methods suffer from poor generalizability to the infrared domain due to fundamental differences in imaging physics. To address this, we propose a Progressive Prompt Fusion Network (PPFN) coupled with a Selective Progressive Training (SPT) strategy, which, for the first time, encodes physical imaging priors as learnable prompt pairs to enable adaptive, joint enhancement of noise, contrast, and blur. Furthermore, we introduce the first high-quality, multi-scenario thermal infrared enhancement benchmark. Extensive experiments demonstrate that our method achieves an 8.76 dB PSNR improvement under complex coupled degradations, significantly outperforming state-of-the-art approaches. The source code is publicly available.

Technology Category

• Computer Vision: Multi-modal Vision
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Natural Language Processing: Prompt Engineering / Prompting

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Security and Privacy: Privacy-enhancing technologies
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

We engage in the relatively underexplored task named thermal infrared image enhancement. Existing infrared image enhancement methods primarily focus on tackling individual degradations, such as noise, contrast, and blurring, making it difficult to handle coupled degradations. Meanwhile, all-in-one enhancement methods, commonly applied to RGB sensors, often demonstrate limited effectiveness due to the significant differences in imaging models. In sight of this, we first revisit the imaging mechanism and introduce a Progressive Prompt Fusion Network (PPFN). Specifically, the PPFN initially establishes prompt pairs based on the thermal imaging process. For each type of degradation, we fuse the corresponding prompt pairs to modulate the model's features, providing adaptive guidance that enables the model to better address specific degradations under single or multiple conditions. In addition, a Selective Progressive Training (SPT) mechanism is introduced to gradually refine the model's handling of composite cases to align the enhancement process, which not only allows the model to remove camera noise and retain key structural details, but also enhancing the overall contrast of the thermal image. Furthermore, we introduce the most high-quality, multi-scenarios infrared benchmark covering a wide range of scenarios. Extensive experiments substantiate that our approach not only delivers promising visual results under specific degradation but also significantly improves performance on complex degradation scenes, achieving a notable 8.76% improvement. Code is available at https://github.com/Zihang-Chen/HM-TIR.

Problem

• Enhancing thermal infrared images with coupled degradations
• Developing adaptive model for single/multiple degradation conditions
• Creating high-quality multi-scenario infrared benchmark dataset

Innovation

• Progressive Prompt Fusion Network modulates features adaptively
• Selective Progressive Training refines composite degradation handling
• High-quality multi-scenario infrared benchmark enables comprehensive evaluation