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.

Self-augmented Unpaired Image Dehazing Via Density and Depth Decomposition

Robust Unpaired Image Dehazing Via Density and Depth Decomposition

Nighttime Image Dehazing by Render.

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

Depth-agnostic Single Image Dehazing

Depth-Aware Unpaired Video Dehazing

Perceiving and Modeling Density is All You Need for Image Dehazing

Autonomous Single-Image Dehazing: Enhancing Local Texture with Haze Density-Aware Image Blending

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

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

Depth Information Assisted Collaborative Mutual Promotion Network for Single Image Dehazing

HEDehazeNet: Unpaired image dehazing via enhanced haze generation

Single image dehazing via cycle-consistent adversarial networks with a multi-scale hybrid encoder-decoder and global correlation loss

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

Single Image Dehazing Using Scene Depth Ordering

Progressive Depth Learning for Single Image Dehazing

Physical-model guided self-distillation network for single image dehazing

Model-Based Single Image Deep Dehazing

Frequency compensated diffusion model for real-scene dehazing

Image Dehazing by an Artificial Image Fusion Method Based on Adaptive Structure Decomposition

Depth-Guided Dehazing Network for Long-Range Aerial Scenes