Abstract:Recent unsupervised image dehazing methods used unpaired real-world training data for enhancing generalization on real-world scenes. However, these methods often require dehazing and rehazing cycles with auxiliary networks for training, resulting in high computational costs and extended training time. In this work, we propose an unsupervised dehazing framework called Contrastive Adaptive Frequency Decomposition Dehazing Network (CAFDD). By incorporating carefully designed network structure and constraints, our CAFDD well avoids additional training overhead and needs only 1.91M parameters. Specifically, we first consider the following insights, including: (1) Haze primarily affects high-frequency components in an image, resulting in blurred edges; (2) Low-frequency components capture the large-scale variations with less susceptibility to haze; and (3) Existing unlearnable frequency decomposition methods such Fourier transform often suffer from information loss, and thus develop the novel PMP (Pointwise convolution-Max pooling-Pointwise convolution) and DAD (Depthwise convolution-Average pooling-Depthwise convolution) blocks to automatically extract high and low-frequency features from input images for accurately estimating transmission map. Then, we propose haze discrimination (HD), a new pretext task for contrastive learning in image dehazing, by forming positive and negative pairs based on haze presence, in order for guiding the network to extract visibility-related features. Last, to get rid of the rehazing cycle and improve training efficiency, we construct a pixel-level constraint, histogram equalization-based texture loss function, which enhances the sharpness and realism of the generated images. Through extensive experiments, we demonstrate the superiority of our CAFDD over the state-of-the-art dehazing approaches on real-world land and overwater images.

Density‐Aware Diffusion Model for Efficient Image Dehazing

Nonuniformly Dehaze Network for Visible Remote Sensing Images

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

DFR-Net: Density Feature Refinement Network for Image Dehazing Utilizing Haze Density Difference

Perceiving and Modeling Density is All You Need for Image Dehazing

High-quality Image Dehazing with Diffusion Model

Single Remote Sensing Image Dehazing Using a Prior-Based Dense Attentive Network

Contrastive adaptive frequency decomposition network guided by haze discrimination for real-world image dehazing

Frequency compensated diffusion model for real-scene dehazing

Haze Relevant Feature Attention Network for Single Image Dehazing

Self-augmented Unpaired Image Dehazing Via Density and Depth Decomposition

RSHazeDiff: A Unified Fourier-aware Diffusion Model for Remote Sensing Image Dehazing

Robust Unpaired Image Dehazing Via Density and Depth Decomposition

LDPC-Net: A Lightweight Detail–Content Progressive Coupled Network for Single-Image Dehazing with Adaptive Feature Extraction Block

A Haze Density Aware Adaptive Perceptual Single Image Haze Removal Algorithm.

Discrete Haze Level Dehazing Network

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

DEA-Net: Single Image Dehazing Based on Detail-Enhanced Convolution and Content-Guided Attention

Dual-path dehazing network with spatial-frequency feature fusion