# I Can See Clearly Now : Image Restoration via De-Raining

## TL;DR

This paper introduces a novel de-raining method that improves image segmentation in rainy conditions by using a new stereo dataset and a trained denoising generator, enhancing performance across multiple datasets.

## Contribution

The paper presents a new stereo dataset with real water droplets and a denoising generator trained on it, demonstrating improved segmentation in rainy images.

## Key findings

- Significant improvement in segmentation accuracy on multiple datasets.
- Effective removal of real water droplets and streaks from images.
- Generalization of de-raining to synthetic water effects.

## Abstract

We present a method for improving segmentation tasks on images affected by adherent rain drops and streaks. We introduce a novel stereo dataset recorded using a system that allows one lens to be affected by real water droplets while keeping the other lens clear. We train a denoising generator using this dataset and show that it is effective at removing the effect of real water droplets, in the context of image reconstruction and road marking segmentation. To further test our de-noising approach, we describe a method of adding computer-generated adherent water droplets and streaks to any images, and use this technique as a proxy to demonstrate the effectiveness of our model in the context of general semantic segmentation. We benchmark our results using the CamVid road marking segmentation dataset, Cityscapes semantic segmentation datasets and our own real-rain dataset, and show significant improvement on all tasks.