RPD-Diff: Region-Adaptive Physics-Guided Diffusion Model for Visibility Enhancement under Dense and Non-Uniform Haze

TL;DR

RPD-Diff is a novel physics-guided diffusion model that adaptively enhances visibility in complex, dense, and non-uniform haze conditions by leveraging physical priors and dynamic denoising strategies.

Contribution

It introduces a region-adaptive, physics-guided diffusion framework with innovative strategies for handling dense and spatially varying haze, outperforming existing methods.

Findings

Achieves state-of-the-art dehazing performance on real-world datasets.

Effectively manages non-uniform haze with dynamic, patch-specific denoising.

Produces high-quality, detail-rich, haze-free images.

Abstract

Single-image dehazing under dense and non-uniform haze conditions remains challenging due to severe information degradation and spatial heterogeneity. Traditional diffusion-based dehazing methods struggle with insufficient generation conditioning and lack of adaptability to spatially varying haze distributions, which leads to suboptimal restoration. To address these limitations, we propose RPD-Diff, a Region-adaptive Physics-guided Dehazing Diffusion Model for robust visibility enhancement in complex haze scenarios. RPD-Diff introduces a Physics-guided Intermediate State Targeting (PIST) strategy, which leverages physical priors to reformulate the diffusion Markov chain by generation target transitions, mitigating the issue of insufficient conditioning in dense haze scenarios. Additionally, the Haze-Aware Denoising Timestep Predictor (HADTP) dynamically adjusts patch-specific denoising timesteps employing a transmission map cross-attention mechanism, adeptly managing non-uniform haze distributions. Extensive experiments across four real-world datasets demonstrate that RPD-Diff achieves state-of-the-art performance in challenging dense and non-uniform haze scenarios, delivering high-quality, haze-free images with superior detail clarity and color fidelity.