Attentive Generative Adversarial Network for Raindrop Removal from a Single Image

TL;DR

This paper introduces an attentive generative adversarial network that effectively removes raindrops from single images by focusing on occluded regions, significantly improving image clarity compared to previous methods.

Contribution

The novel integration of visual attention into both generative and discriminative networks for raindrop removal enhances focus on occluded areas, advancing image restoration techniques.

Findings

Outperforms state-of-the-art methods quantitatively.

Achieves superior qualitative image restoration.

Effectively learns raindrop regions and surroundings during training.

Abstract

Raindrops adhered to a glass window or camera lens can severely hamper the visibility of a background scene and degrade an image considerably. In this paper, we address the problem by visually removing raindrops, and thus transforming a raindrop degraded image into a clean one. The problem is intractable, since first the regions occluded by raindrops are not given. Second, the information about the background scene of the occluded regions is completely lost for most part. To resolve the problem, we apply an attentive generative network using adversarial training. Our main idea is to inject visual attention into both the generative and discriminative networks. During the training, our visual attention learns about raindrop regions and their surroundings. Hence, by injecting this information, the generative network will pay more attention to the raindrop regions and the surrounding structures, and the discriminative network will be able to assess the local consistency of the restored regions. This injection of visual attention to both generative and discriminative networks is the main contribution of this paper. Our experiments show the effectiveness of our approach, which outperforms the state of the art methods quantitatively and qualitatively.