The Devil is in the Darkness: Diffusion-Based Nighttime Dehazing Anchored in Brightness Perception

2025-06-03

AI Summary

Nighttime image dehazing struggles to restore authentic daytime brightness due to the absence of explicit diurnal brightness mapping in existing datasets and the lack of explicit daytime illumination priors in current models. To address this, we propose the first end-to-end brightness-aware diffusion framework. Our method comprises: (1) a novel synthetic data pipeline explicitly modeling the diurnal brightness mapping relationship; and (2) a brightness-guided diffusion architecture integrating pre-trained diffusion priors, physics-driven brightness constraints, and joint optimization. Evaluated on multiple nighttime dehazing benchmarks, our approach achieves significant improvements in PSNR and SSIM, enabling more natural daylight brightness restoration, superior detail fidelity, and strong generalization. Both qualitative visual realism and quantitative metrics surpass state-of-the-art methods.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Natural Language Processing: Language Grounding & Multi-modal NLP

Application Category

• Security and Privacy: Data transparency and provenance
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

While nighttime image dehazing has been extensively studied, converting nighttime hazy images to daytime-equivalent brightness remains largely unaddressed. Existing methods face two critical limitations: (1) datasets overlook the brightness relationship between day and night, resulting in the brightness mapping being inconsistent with the real world during image synthesis; and (2) models do not explicitly incorporate daytime brightness knowledge, limiting their ability to reconstruct realistic lighting. To address these challenges, we introduce the Diffusion-Based Nighttime Dehazing (DiffND) framework, which excels in both data synthesis and lighting reconstruction. Our approach starts with a data synthesis pipeline that simulates severe distortions while enforcing brightness consistency between synthetic and real-world scenes, providing a strong foundation for learning night-to-day brightness mapping. Next, we propose a restoration model that integrates a pre-trained diffusion model guided by a brightness perception network. This design harnesses the diffusion model's generative ability while adapting it to nighttime dehazing through brightness-aware optimization. Experiments validate our dataset's utility and the model's superior performance in joint haze removal and brightness mapping.

Problem

• Convert nighttime hazy images to daytime brightness accurately
• Address inconsistent brightness mapping in existing datasets
• Enhance lighting reconstruction with daytime brightness knowledge

Innovation

• Brightness-consistent synthetic data pipeline
• Diffusion model with brightness perception
• Night-to-day brightness mapping optimization