Clearing the Skies: A deep network architecture for single-image rain removal

TL;DR

This paper presents DerainNet, a deep CNN architecture trained on synthetic data to effectively remove rain streaks from images, outperforming existing methods in quality and speed.

Contribution

The paper introduces DerainNet, a novel CNN-based approach trained on detail layers and enhanced with image processing knowledge for improved rain removal.

Findings

Outperforms state-of-the-art rain removal methods

Operates faster after training

Effectively generalizes from synthetic to real images

Abstract

We introduce a deep network architecture called DerainNet for removing rain streaks from an image. Based on the deep convolutional neural network (CNN), we directly learn the mapping relationship between rainy and clean image detail layers from data. Because we do not possess the ground truth corresponding to real-world rainy images, we synthesize images with rain for training. In contrast to other common strategies that increase depth or breadth of the network, we use image processing domain knowledge to modify the objective function and improve deraining with a modestly-sized CNN. Specifically, we train our DerainNet on the detail (high-pass) layer rather than in the image domain. Though DerainNet is trained on synthetic data, we find that the learned network translates very effectively to real-world images for testing. Moreover, we augment the CNN framework with image enhancement to improve the visual results. Compared with state-of-the-art single image de-raining methods, our method has improved rain removal and much faster computation time after network training.