Abstract:Supervised dehazing models, trained on synthetic hazy-clean image pairs, often face a notable decline in performance when applied to real-world scenes. Consequently, CycleGAN-based unpaired dehazing methods are proposed to improve the model’s generalization. One successful approach among these methods involves decomposing the physical properties of the atmospheric scattering model (ASM). However, estimating physical properties individually from input images is difficult without supervised labels, which ignores the semantic consistency between different physical regions. We claim semantic region information can offer additional geometric spatial constraints for estimating physical properties, as natural images can be divided into regions with similar scene depths. Motivated by this, we propose a novel generalized and realistic unpaired image dehazing framework via region-aware physical constraints (RPC-Dehaze). Our approach utilizes fine-grained semantic region maps from the Segment Anything Model (SAM) in a specially designed region prompt enhancement module. This enables the dehazing and hazing cyclic networks to learn region-aware physical constraints, leading to accurate estimation of haze imaging physical properties. In contrast to existing unpaired methods that treat dehazing and hazing networks equally, we incorporate Retinex theory into the hazing network, allowing it to learn diverse illumination effects in different regions. We adaptively refine the Retinex-based illumination component, resulting in more realistic hazy images. To further facilitate unsupervised learning in our framework, we propose a physical consensual contrastive regularization to ensure compact representation constraints in the latent feature space. Extensive experiments on synthetic and real image datasets show our method surpasses state-of-the-art unpaired dehazing methods in both effectiveness and generalization capability.

Towards Generalization on Real Domain for Single Image Dehazing Via Meta-Learning

Dehaze on Small-Scale Datasets Via Self-Supervised Learning

From Coarse to Fine: A Two-Stage Network for Dense Haze removal

Semi-Supervised Domain Alignment Learning for Single Image Dehazing

MsDA: Multi-scale Domain Adaptation Dehazing Network

DADRnet: Cross-domain Image Dehazing Via Domain Adaptation and Disentangled Representation.

Mutual Learning for Domain Adaptation: Self-distillation Image Dehazing Network with Sample-cycle

Generative Adversarial and Self-Supervised Dehazing Network

Towards Domain Invariant Single Image Dehazing

Source-Free Domain Adaptation for Real-World Image Dehazing

From Synthetic to Real: Image Dehazing Collaborating with Unlabeled Real Data

Semi-supervised Single-Image Dehazing Network via Disentangled Meta-Knowledge

Accurate and lightweight dehazing via multi-receptive-field non-local network and novel contrastive regularization

Robust Real-World Image Dehazing Via Knowledge Guided Conditional Diffusion Model Finetuning

Toward Generalized and Realistic Unpaired Image Dehazing Via Region-Aware Physical Constraints

Domain Prompt Learning Framework for Real Image Dehazing

Advancing Real-World Image Dehazing: Perspective, Modules, and Training

Remote sensing image dehazing based on data blending and Laplace network

Two-step image dehazing with intra-domain and inter-domain adaptation

Single image mixed dehazing method based on numerical iterative model and DehazeNet

Discrete Haze Level Dehazing Network