API: Empowering Generalizable Real-World Image Dehazing via Adaptive Patch Importance Learning

TL;DR

This paper introduces a novel adaptive patch importance learning framework for real-world image dehazing, combining data augmentation, density-aware haze removal, and a contrastive loss to improve generalization and achieve state-of-the-art results.

Contribution

The paper proposes a new API framework with modules for realistic haze data generation and adaptive haze removal, along with a novel contrastive loss for better detail preservation.

Findings

Achieves state-of-the-art performance on real-world haze benchmarks.

Demonstrates strong generalization across diverse haze conditions.

Improves visual quality and quantitative metrics over existing methods.

Abstract

Real-world image dehazing is a fundamental yet challenging task in low-level vision. Existing learning-based methods often suffer from significant performance degradation when applied to complex real-world hazy scenes, primarily due to limited training data and the intrinsic complexity of haze density distributions.To address these challenges, we introduce a novel Adaptive Patch Importance-aware (API) framework for generalizable real-world image dehazing. Specifically, our framework consists of an Automatic Haze Generation (AHG) module and a Density-aware Haze Removal (DHR) module. AHG provides a hybrid data augmentation strategy by generating realistic and diverse hazy images as additional high-quality training data. DHR considers hazy regions with varying haze density distributions for generalizable real-world image dehazing in an adaptive patch importance-aware manner. To alleviate the ambiguity of the dehazed image details, we further introduce a new Multi-Negative Contrastive Dehazing (MNCD) loss, which fully utilizes information from multiple negative samples across both spatial and frequency domains. Extensive experiments demonstrate that our framework achieves state-of-the-art performance across multiple real-world benchmarks, delivering strong results in both quantitative metrics and qualitative visual quality, and exhibiting robust generalization across diverse haze distributions.