Global Information Interactive of ViT for Rain Removal on Single Images

doi:10.1109/CAC59555.2023.10450755

IR > 外国语学院

	Global Information Interactive of ViT for Rain Removal on Single Images
	Li, Ce; Zhao, Shutian; Huang, Fan; Ma, Pengfei; Ma, Lin; Chen, Huizhong
	2023
会议名称	2023 China Automation Congress, CAC 2023
会议录名称	Proceedings - 2023 China Automation Congress, CAC 2023
页码	8411-8416
会议日期	November 17, 2023 - November 19, 2023
会议地点	Chongqing, China
出版者	Institute of Electrical and Electronics Engineers Inc.
摘要	In adverse weather conditions, the quality of outdoor captured images is significantly compromised, primarily due to the presence of rain streaks that severely degrade image clarity and consequently disrupt image recognition and analysis. The spatial variations of rain streaks within individual rainy images pose a considerable challenge for their removal. Despite the notable achievements of Convolutional Neural Network (CNN)-based methods in recent years, their limited receptive fields and lack of adaptability to input content make it difficult for them to cope with real-world scenarios and restore high-quality rain-free images with accurate structural details. To address this issue, we propose an image restoration model based on the Vision Transformer (ViT) architecture. Specifically, our model first employs a U - ViT to capture contextual information, which is then fused with a high-resolution branch that preserves local details. At each stage, attention mechanisms are introduced to reweight local features, leading to the design of a novel information interaction pattern. The model achieves accelerated convergence by performing global information interaction during rain removal, resulting in image restoration that closely resembles the reference images. Furthermore, to mitigate the loss of fine grained details, we introduce a detail fusion module. The resulting tightly interconnected hierarchical structure is referred to as GIVTNet. Experimental results demonstrate that the proposed algorithm yields substantial performance improvements across multiple synthetic and real-world datasets. © 2023 IEEE.
关键词	Computer vision Convolutional neural networks Image recognition Image reconstruction Restoration Adverse weather Condition Detail fusion Global informations Image clarity Information interaction Rain removals Single image deraining Single images Vision transformer
DOI	10.1109/CAC59555.2023.10450755
收录类别	EI
语种	英语
EI入藏号	20241515852307
EI主题词	Rain
EI分类号	443.3 Precipitation ; 723.5 Computer Applications ; 741.2 Vision
原始文献类型	Conference article (CA)
引用统计	无
文献类型	会议论文
条目标识符	https://ir.lut.edu.cn/handle/2XXMBERH/170545
专题	外国语学院
通讯作者	Li, Ce
作者单位	School of Electrical Engineering and Information Engineering, Lanzhou University of Technology, Lanzhou, China
第一作者单位	兰州理工大学
通讯作者单位	兰州理工大学
推荐引用方式 GB/T 7714	Li, Ce,Zhao, Shutian,Huang, Fan,et al. Global Information Interactive of ViT for Rain Removal on Single Images[C]:Institute of Electrical and Electronics Engineers Inc.,2023:8411-8416.