PRISM: Rethinking Scattered Atmosphere Reconstruction as a Unified Understanding and Generation Model for Real-world Dehazing

TL;DR

PRISM introduces a physically structured framework for real-world image dehazing that jointly reconstructs scenes and scattering variables, employing novel synthesis and adaptation techniques to handle complex haze conditions.

Contribution

It presents a unified model that combines atmospheric scattering reconstruction with adaptive learning strategies to improve real-world dehazing performance.

Findings

Achieves state-of-the-art results on real-world dehazing benchmarks.

Effectively handles non-uniform haze and mixed-light conditions.

Demonstrates robustness in unpaired real-world scenarios.

Abstract

Real-world image dehazing (RID) aims to remove haze induced degradation from real scenes. This task remains challenging due to non-uniform haze distribution, spatially varying illumination from multiple light sources, and the scarcity of paired real hazy-clean data. In PRISM, we propose Proximal Scattered Atmosphere Reconstruction (PSAR), a physically structured framework that jointly reconstructs the clear scene and scattering variables under the atmospheric scattering model, thereby improving reliability in complex regions and mixed-light conditions. To bridge the synthetic-to-real gap, we design an online non-uniform haze synthesis pipeline and a Selective Self-distillation Adaptation scheme for unpaired real-world scenarios, which enables the model to selectively learn from high-quality perceptual targets while leveraging its intrinsic scattering understanding to audit residual haze and guide self-refinement. Extensive experiments on real-world benchmarks demonstrate that PRISM achieves state-of-the-art performance on RID tasks.