Single Image Rain Streak Removal Using Harris Corner Loss and R-CBAM Network

TL;DR

This paper introduces a novel rain streak removal network that uses Harris Corner Loss and R-CBAM attention modules to better preserve details and improve rain removal performance in single images.

Contribution

The study proposes a new network architecture with Corner Loss and R-CBAM modules, enhancing detail preservation and focus on rain-affected regions during single-image rain streak removal.

Findings

Achieves higher PSNR on Rain100L and Rain100H datasets.

Outperforms previous methods in rain streak removal quality.

Effectively preserves object boundaries and textures.

Abstract

The problem of single-image rain streak removal goes beyond simple noise suppression, requiring the simultaneous preservation of fine structural details and overall visual quality. In this study, we propose a novel image restoration network that effectively constrains the restoration process by introducing a Corner Loss, which prevents the loss of object boundaries and detailed texture information during restoration. Furthermore, we propose a Residual Convolutional Block Attention Module (R-CBAM) Block into the encoder and decoder to dynamically adjust the importance of features in both spatial and channel dimensions, enabling the network to focus more effectively on regions heavily affected by rain streaks. Quantitative evaluations conducted on the Rain100L and Rain100H datasets demonstrate that the proposed method significantly outperforms previous approaches, achieving a PSNR of 33.29 dB on Rain100L and 26.16 dB on Rain100H.