Stereo superpixel: An iterative framework based on parallax consistency and collaborative optimization

doi:10.1016/j.ins.2020.12.031

	Stereo superpixel: An iterative framework based on parallax consistency and collaborative optimization
	Li, Hua 1,2,3; Cong, Runmin 4,5; Kwong, Sam 1; Chen, Chuanbo 2; Xu, Qianqian 6; Li, Chongyi 7
	2021-05-01
发表期刊	INFORMATION SCIENCES
ISSN	0020-0255
卷号	556 页码:209-222
摘要	Stereo superpixel segmentation aims to obtain the superpixel segmentation results of the left and right views more cooperatively and consistently, rather than simply performing independent segmentation directly. Thus, the correspondence between two views should be reasonably modeled and fully considered. In this paper, we propose a left-right interactive optimization framework for stereo superpixel segmentation. Considering the disparity in stereo image pairs, we first divide the images into paired region and non-paired region, and propose a collaborative optimization scheme to coordinately refine the matched superpixels of the left and right views in an interactive manner. This is, to the best of our knowledge, the first attempt to generate stereo superpixels considering the parallax consistency. Quantitative and qualitative experiments demonstrate that the proposed framework achieves superior performance in terms of consistency and accuracy compared with single-image superpixel segmentation. (C) 2020 Elsevier Inc. All rights reserved.
关键词	Stereo superpixel Superpixel segmentation Parallax consistency Collaborative optimization
DOI	10.1016/j.ins.2020.12.031
收录类别	SCI
语种	英语
资助项目	Key Project of Science and Technology Innovation 2030 - Ministry of Science and Technology of China[2018AAA0101301] ; Natural Science Foundation of China[61772344] ; Natural Science Foundation of China[62002014] ; Hong Kong RGC General Research Funds[9042816] ; Hong Kong RGC General Research Funds[CityU 11209819] ; Beijing Nova Program[Z201100006820016] ; Fundamental Research Funds for the Central Universities[2019RC039] ; Elite Scientist Sponsorship Program by Beijing Association for Science and Technology ; Hong Kong Scholars Program ; China Postdoctoral Science Foundation[2020T130050] ; China Postdoctoral Science Foundation[2019M660438]
WOS研究方向	Computer Science
WOS类目	Computer Science, Information Systems
WOS记录号	WOS:000626586900013
出版者	ELSEVIER SCIENCE INC
引用统计	被引频次：8[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://119.78.100.204/handle/2XEOYT63/16795
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Cong, Runmin; Kwong, Sam
作者单位	1.City Univ Hong Kong, Dept Comp Sci, Kowloon, Hong Kong, Peoples R China 2.Huazhong Univ Sci & Technol, Sch Software Engn, Wuhan 430074, Peoples R China 3.Hainan Univ, Sch Comp Sci & Cyberspace Secur, Haikou 570228, Hainan, Peoples R China 4.Beijing Jiaotong Univ, Inst Informat Sci, Beijing 100044, Peoples R China 5.Beijing Jiaotong Univ, Beijing Key Lab Adv Informat Sci & Network Techno, Beijing 100044, Peoples R China 6.Chinese Acad Sci, Inst Comp Technol, Key Lab Intelligent Informat Proc, Beijing 100864, Peoples R China 7.Nanyang Technol Univ, Singapore 639798, Singapore
推荐引用方式 GB/T 7714	Li, Hua,Cong, Runmin,Kwong, Sam,et al. Stereo superpixel: An iterative framework based on parallax consistency and collaborative optimization[J]. INFORMATION SCIENCES,2021,556:209-222.
APA	Li, Hua,Cong, Runmin,Kwong, Sam,Chen, Chuanbo,Xu, Qianqian,&Li, Chongyi.(2021).Stereo superpixel: An iterative framework based on parallax consistency and collaborative optimization.INFORMATION SCIENCES,556,209-222.
MLA	Li, Hua,et al."Stereo superpixel: An iterative framework based on parallax consistency and collaborative optimization".INFORMATION SCIENCES 556(2021):209-222.