A^2Net: Adjacent Aggregation Networks for Image Raindrop Removal

TL;DR

A^2Net is a lightweight neural network that effectively removes raindrops from images by using adjacent aggregation and luminance-focused learning, achieving high quality results with fewer parameters.

Contribution

The paper introduces a novel adjacent aggregation architecture and luminance-focused training for raindrop removal, reducing model complexity while improving performance.

Findings

State-of-the-art raindrop removal performance

Significant reduction in model parameters

Effective feature fusion through adjacent aggregation

Abstract

Existing methods for single images raindrop removal either have poor robustness or suffer from parameter burdens. In this paper, we propose a new Adjacent Aggregation Network (A^2Net) with lightweight architectures to remove raindrops from single images. Instead of directly cascading convolutional layers, we design an adjacent aggregation architecture to better fuse features for rich representations generation, which can lead to high quality images reconstruction. To further simplify the learning process, we utilize a problem-specific knowledge to force the network focus on the luminance channel in the YUV color space instead of all RGB channels. By combining adjacent aggregating operation with color space transformation, the proposed A^2Net can achieve state-of-the-art performances on raindrop removal with significant parameters reduction.