A Comprehensive Survey and Taxonomy on Single Image Dehazing Based on Deep Learning

TL;DR

This paper provides a comprehensive survey of deep learning-based single image dehazing methods, categorizing approaches, analyzing datasets and metrics, and discussing future challenges in the field.

Contribution

It offers a detailed taxonomy and comparison of supervised, semi-supervised, and unsupervised dehazing algorithms, along with experimental evaluations.

Findings

Deep learning methods significantly improve dehazing quality.

Various datasets and evaluation metrics are analyzed.

Future research challenges are identified.

Abstract

With the development of convolutional neural networks, hundreds of deep learning based dehazing methods have been proposed. In this paper, we provide a comprehensive survey on supervised, semi-supervised, and unsupervised single image dehazing. We first discuss the physical model, datasets, network modules, loss functions, and evaluation metrics that are commonly used. Then, the main contributions of various dehazing algorithms are categorized and summarized. Further, quantitative and qualitative experiments of various baseline methods are carried out. Finally, the unsolved issues and challenges that can inspire the future research are pointed out. A collection of useful dehazing materials is available at \url{https://github.com/Xiaofeng-life/AwesomeDehazing}.