PHATNet: A Physics-guided Haze Transfer Network for Domain-adaptive Real-world Image Dehazing

TL;DR

PHATNet is a novel physics-guided network that transfers haze patterns from unseen domains to improve real-world image dehazing through domain adaptation and specialized loss functions.

Contribution

It introduces a physics-guided haze transfer approach with new loss functions for effective domain adaptation in image dehazing.

Findings

Significantly improves dehazing performance on real-world datasets.

Outperforms state-of-the-art models in benchmark tests.

Enhances model robustness to unseen haze patterns.

Abstract

Image dehazing aims to remove unwanted hazy artifacts in images. Although previous research has collected paired real-world hazy and haze-free images to improve dehazing models' performance in real-world scenarios, these models often experience significant performance drops when handling unseen real-world hazy images due to limited training data. This issue motivates us to develop a flexible domain adaptation method to enhance dehazing performance during testing. Observing that predicting haze patterns is generally easier than recovering clean content, we propose the Physics-guided Haze Transfer Network (PHATNet) which transfers haze patterns from unseen target domains to source-domain haze-free images, creating domain-specific fine-tuning sets to update dehazing models for effective domain adaptation. Additionally, we introduce a Haze-Transfer-Consistency loss and a Content-Leakage Loss to enhance PHATNet's disentanglement ability. Experimental results demonstrate that PHATNet significantly boosts state-of-the-art dehazing models on benchmark real-world image dehazing datasets.