Accurate blind deblurring using salientpatch-based prior for large-size images

doi:10.1007/s11042-018-6009-2

CORC > 自动化研究所 > 中国科学院自动化研究所 > 多媒体计算与图形学团队

	Accurate blind deblurring using salientpatch-based prior for large-size images
	Ma, Chengcheng 1,2,3; Zhang, Jiguang 4; Xu, Shibiao3 ; Meng, Weiliang3 ; Xi, Runping 1; Kumar, G. Hemantha 4; Zhang, Xiaopeng3
刊名	MULTIMEDIA TOOLS AND APPLICATIONS
	2018-11-01
卷号	77 期号:21 页码:28077-28100
关键词	Deblurring SalientPatch Kernel estimation Segmentation Large-size
ISSN号	1380-7501
DOI	10.1007/s11042-018-6009-2
通讯作者	Xu, Shibiao(shibiao.xu@ia.ac.cn) ; Zhang, Xiaopeng(xiaopeng.zhang@ia.ac.cn)
英文摘要	The full-image based kernel estimation strategy is usually susceptible by the smooth and fine-scale background regions impacting and it is time-consuming for large-size image deblurring. Since not all the pixels in the blurred image are informative and it is frequent to restore human-interested objects in the foreground rather than background, we propose a novel concept "SalientPatch" to denote informative regions for better blur kernel estimation without user guidance by computing three cues (objectness probability, structure richness and local contrast). Although these cues are not new, it is innovative to integrate and complement each other in motion blur restoration. Experiments demonstrate that our SalientPatch-based deblurring algorithm can significantly speed up the kernel estimation and guarantee high-quality recovery for large-size blurry images as well.
资助项目	National Natural Science Foundation of China[61620106003] ; National Natural Science Foundation of China[61671451] ; National Natural Science Foundation of China[61572405] ; National Natural Science Foundation of China[61502490] ; National Natural Science Foundation of China[61571439] ; National Natural Science Foundation of China[61771026] ; Open Projects Program of National Laboratory of Pattern Recognition[201600038] ; Independent Research Project of National Laboratory of Pattern Recognition[Z-2018005] ; Independent Research Project of National Laboratory of Pattern Recognition[6140001010207]
WOS关键词	SALIENCY DETECTION ; DECONVOLUTION
WOS研究方向	Computer Science ; Engineering
语种	英语
出版者	SPRINGER
WOS记录号	WOS:000446601500015
资助机构	National Natural Science Foundation of China ; Open Projects Program of National Laboratory of Pattern Recognition ; Independent Research Project of National Laboratory of Pattern Recognition
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/23040]
专题	多媒体计算与图形学团队
通讯作者	Xu, Shibiao; Zhang, Xiaopeng
作者单位	1.Northwestern Polytech Univ, Sch Comp Sci, Xian, Shaanxi, Peoples R China 2.Univ Chinese Acad Sci, Beijing, Peoples R China 3.Chinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing, Peoples R China 4.Univ Mysore, Dept Comp Sci, Mysore, Karnataka, India
推荐引用方式 GB/T 7714	Ma, Chengcheng,Zhang, Jiguang,Xu, Shibiao,et al. Accurate blind deblurring using salientpatch-based prior for large-size images[J]. MULTIMEDIA TOOLS AND APPLICATIONS,2018,77(21):28077-28100.
APA	Ma, Chengcheng.,Zhang, Jiguang.,Xu, Shibiao.,Meng, Weiliang.,Xi, Runping.,...&Zhang, Xiaopeng.(2018).Accurate blind deblurring using salientpatch-based prior for large-size images.MULTIMEDIA TOOLS AND APPLICATIONS,77(21),28077-28100.
MLA	Ma, Chengcheng,et al."Accurate blind deblurring using salientpatch-based prior for large-size images".MULTIMEDIA TOOLS AND APPLICATIONS 77.21(2018):28077-28100.