Single Image Deraining Integrating Physics Model and Density-Oriented Conditional GAN Refinement

doi:10.1109/LSP.2021.3095613

Single Image Deraining Integrating Physics Model and Density-Oriented Conditional GAN Refinement

2021-01-01

8

OA

AI

DOI：10.1109/LSP.2021.3095613

Original

Original request for help

Save

Abstract

En 中文

Although advanced single image deraining methods have been proposed, their generalization ability to real-world images is usually limited, especially when dealing with rain patterns of different densities, shapes, and directions. In order to improve the robustness and generalization of these deraining methods, we propose a novel density-aware single image deraining method with gated multi-scale feature fusion, which consists of two stages. In the first stage, a sophisticated physics model is leveraged for initial deraining and a network branch is utilized for rain density estimation to guide the subsequent refinement. The second stage of model-independent refinement is realized using conditional Generative Adversarial Network (cGAN), attempting to eliminate artifacts and improve the restoration quality. Extensive experiments have been conducted on the representative synthetic rain datasets and real rain scenes, demonstrating the superiority of our method in terms of effectiveness and generalization ability, which outperforms the state-of-the-arts.

Keywords:

Rain

Training

Atmospheric modeling

Logic gates

Image restoration

Feature extraction

Signal processing algorithms

Single image deraining

cGAN

gated fusion

physics model

rain density classification

Single Image Deraining Integrating Physics Model and Density-Oriented Conditional GAN Refinement

Abstract

Journal

Organization