Semi-DerainGAN: A New Semi-supervised Single Image Deraining Network

TL;DR

Semi-DerainGAN is a semi-supervised GAN-based network that effectively removes rain streaks from real images by leveraging both synthetic and real rainy images, improving generalization over traditional supervised methods.

Contribution

The paper introduces Semi-DerainGAN, a semi-supervised deraining network with a shared rain streak learner and paired discriminator, along with a new real-world rainy image dataset.

Findings

Achieves competitive deraining performance on public datasets.

Effectively handles real images with diverse rain streaks.

Outperforms purely supervised models on real-world data.

Abstract

Removing the rain streaks from single image is still a challenging task, since the shapes and directions of rain streaks in the synthetic datasets are very different from real images. Although supervised deep deraining networks have obtained impressive results on synthetic datasets, they still cannot obtain satisfactory results on real images due to weak generalization of rain removal capacity, i.e., the pre-trained models usually cannot handle new shapes and directions that may lead to over-derained/under-derained results. In this paper, we propose a new semi-supervised GAN-based deraining network termed Semi-DerainGAN, which can use both synthetic and real rainy images in a uniform network using two supervised and unsupervised processes. Specifically, a semi-supervised rain streak learner termed SSRML sharing the same parameters of both processes is derived, which makes the real images contribute more rain streak information. To deliver better deraining results, we design a paired discriminator for distinguishing the real pairs from fake pairs. Note that we also contribute a new real-world rainy image dataset Real200 to alleviate the difference between the synthetic and real image do-mains. Extensive results on public datasets show that our model can obtain competitive performance, especially on real images.