CFNet: Conditional filter learning with dynamic noise estimation for real image denoising

doi:10.1016/j.knosys.2023.111320

	CFNet: Conditional filter learning with dynamic noise estimation for real image denoising
	Zuo, Yifan 1; Yao, Wenhao 1; Zeng, Yifeng 2; Xie, Jiacheng 1; Fang, Yuming 1; Huang, Yan 3; Jiang, Wenhui 1
发表期刊	KNOWLEDGE-BASED SYSTEMS
ISSN	0950-7051
	2024-01-25
卷号	284 页码:12
通讯作者	Fang, Yuming(fa0001ng@e.ntu.edu.sg)
摘要	A mainstream type of the state of the arts (SOTAs) based on convolutional neural network (CNN) for real image denoising contains two sub-problems, i.e., noise estimation and non-blind denoising. This paper considers real noise approximated by heteroscedastic Gaussian/Poisson-Gaussian distributions with in-camera signal processing pipelines. The related works always exploit the estimated noise prior via channel-wise concatenation followed by a convolutional layer with spatially sharing kernels. Due to the variable modes of noise strength and frequency details of all feature positions, this design cannot adaptively tune the corresponding denoising patterns. To address this problem, we propose a novel conditional filter in which the optimal kernels for different feature positions can be adaptively inferred by local features from the image and the noise map. Also, we bring the thought that alternatively performs noise estimation and non-blind denoising into CNN structure, which continuously updates noise prior to guide the iterative feature denoising. In addition, according to the property of heteroscedastic Gaussian distribution, a novel affine transform block is designed to predict the stationary noise component and the signal-dependent noise component. Compared with SOTAs, extensive experiments are conducted on five synthetic datasets and four real datasets, which shows the improvement of the proposed CFNet. The code and models are available via https://github.com/WenhaoYao/CFNet/.
关键词	Image denoising Noise estimation Conditional filter Affine transform
DOI	10.1016/j.knosys.2023.111320
关键词[WOS]	NETWORK
收录类别	SCI
语种	英语
资助项目	National Natural Science Foundation of China[62271237] ; National Natural Science Foundation of China[62132006] ; Natural Science Foundation of Jiangxi Province[20224ACB212005] ; Natural Science Foundation of Jiangxi Province[20223AEI91002] ; Natural Science Foundation of Jiangxi Province[20224BAB212010] ; Double Thousand Plan of Jiangxi Province[jxsq2019101076]
项目资助者	National Natural Science Foundation of China ; Natural Science Foundation of Jiangxi Province ; Double Thousand Plan of Jiangxi Province
WOS研究方向	Computer Science
WOS类目	Computer Science, Artificial Intelligence
WOS记录号	WOS:001149545300001
出版者	ELSEVIER
引用统计	被引频次：2[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://ir.ia.ac.cn/handle/173211/55447
专题	多模态人工智能系统全国重点实验室
通讯作者	Fang, Yuming
作者单位	1.Jiangxi Univ Finance & Econ, Sch Informat Management, 665 Yuping West St, Nanchang 330031, Jiangxi, Peoples R China 2.Northumbria Univ, Dept Comp & Informat Sci, 110 Middlesex St,330031, London, England 3.Chinese Acad Sci, Inst Automat, 95 Zhongguancun East Rd, Beijing 100190, Peoples R China
推荐引用方式 GB/T 7714	Zuo, Yifan,Yao, Wenhao,Zeng, Yifeng,et al. CFNet: Conditional filter learning with dynamic noise estimation for real image denoising[J]. KNOWLEDGE-BASED SYSTEMS,2024,284:12.
APA	Zuo, Yifan.,Yao, Wenhao.,Zeng, Yifeng.,Xie, Jiacheng.,Fang, Yuming.,...&Jiang, Wenhui.(2024).CFNet: Conditional filter learning with dynamic noise estimation for real image denoising.KNOWLEDGE-BASED SYSTEMS,284,12.
MLA	Zuo, Yifan,et al."CFNet: Conditional filter learning with dynamic noise estimation for real image denoising".KNOWLEDGE-BASED SYSTEMS 284(2024):12.