MCGKT-Net: Multi-level Context Gating Knowledge Transfer Network for Single Image Deraining

TL;DR

The paper introduces MCGKT-Net, a multi-scale deep learning framework with internal and external knowledge transfer modules, significantly improving single image deraining performance on benchmark datasets.

Contribution

It proposes a novel multi-scale learning network with internal and external knowledge transfer modules for enhanced rain streak removal.

Findings

Achieves superior results on Rain100H, Rain100L, and Rain800 datasets.

Demonstrates the effectiveness of multi-scale context gating in deraining.

Outperforms state-of-the-art methods in quantitative metrics.

Abstract

Rain streak removal in a single image is a very challenging task due to its ill-posed nature in essence. Recently, the end-to-end learning techniques with deep convolutional neural networks (DCNN) have made great progress in this task. However, the conventional DCNN-based deraining methods have struggled to exploit deeper and more complex network architectures for pursuing better performance. This study proposes a novel MCGKT-Net for boosting deraining performance, which is a naturally multi-scale learning framework being capable of exploring multi-scale attributes of rain streaks and different semantic structures of the clear images. In order to obtain high representative features inside MCGKT-Net, we explore internal knowledge transfer module using ConvLSTM unit for conducting interaction learning between different layers and investigate external knowledge transfer module for leveraging the knowledge already learned in other task domains. Furthermore, to dynamically select useful features in learning procedure, we propose a multi-scale context gating module in the MCGKT-Net using squeeze-and-excitation block. Experiments on three benchmark datasets: Rain100H, Rain100L, and Rain800, manifest impressive performance compared with state-of-the-art methods.