HNDiff: Haze-Noise Diffusion for Image Dehazing

2026-08-11

AI Summary

This work addresses the limitation of existing diffusion-based dehazing methods, which disregard the physical formation mechanism of haze and merely reconstruct images from Gaussian noise, thereby constraining restoration performance. To overcome this, the authors propose HNDiff, a novel framework that embeds the atmospheric scattering model as an inductive bias into the diffusion process: during the forward pass, haze and noise are jointly injected, while the reverse pass simultaneously performs dehazing and denoising to achieve physically consistent image recovery. The method introduces a haze-aware noise scheduler that adaptively modulates noise intensity according to haze density and further presents Latent HNDiff to enhance off-the-shelf dehazing networks. Extensive experiments demonstrate that HNDiff significantly boosts the performance of mainstream backbone architectures across multiple benchmarks, achieving state-of-the-art results.

Abstract

Existing diffusion-based methods have recently made significant progress in image dehazing. However, they typically neglect the physics of haze formation and reconstruct clean images from pure Gaussian noise, thereby limiting their restoration potential. To address this issue, we propose Haze-Noise Diffusion (HNDiff), a novel diffusion framework that embeds the atmospheric scattering model as an inductive bias. By grounding diffusion in physical principles, HNDiff ensures that the restoration aligns more closely with underlying mechanisms of haze formation. In its forward process, we introduce joint haze-noise diffusion with a haze-aware noise scheduler, which progressively adds both haze and noise to an image. Essentially, the scheduler adapts noise levels according to haze density, meaning that regions with heavier haze receive stronger noise injection to encourage content generation, while clearer regions receive lighter noise to better preserve details, which directly links the forward degradation process with the physics of haze. In the reverse process, we then derive a physically consistent dehazing-denoising process that simultaneously removes haze and noise to restore a clean image in a manner aligned with the forward degradation process. To further enhance practicality, we propose Latent HNDiff, which compiles clean latent priors that can be seamlessly integrated into existing dehazing networks to boost performance. Extensive experiments show that our work significantly improves leading dehazing backbones and achieves state-of-the-art results on benchmark datasets. The project page is available at https://jin-ting-he.github.io/HNDiff .

Problem

• image dehazing
• diffusion models
• atmospheric scattering
• haze formation physics
• noise modeling

Innovation

• Haze-Noise Diffusion
• atmospheric scattering model
• haze-aware noise scheduler
• physically consistent dehazing
• latent prior integration