Efficient and robust large-scale structure-from-motion via track selection and camera prioritization

doi:10.1016/j.isprsjprs.2019.08.005

	Efficient and robust large-scale structure-from-motion via track selection and camera prioritization
	Cui, Hainan1 ; Shen, Shuhan1,2 ; Gao, Wei1,2 ; Liu, Hongmin 4; Wang, Zhiheng 3
刊名	ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING
	2019-10-01
卷号	156 页码:202-214
关键词	Image-based modeling Structure-from-motion Track selection Rotation averaging Camera prioritization
ISSN号	0924-2716
DOI	10.1016/j.isprsjprs.2019.08.005
通讯作者	Cui, Hainan(hncui@nlpr.ia.ac.cn) ; Liu, Hongmin(hmliu_82@163.com)
英文摘要	Incremental Structure-from-Motion (SfM) techniques have exhibited superior practicability in many recent studies; however, efficiency and robustness remain key challenges for these techniques. In this work, we propose a new incremental SfM method that overcomes these problems in a united framework that contains two iteration loops. The inner loop is a track selection loop, where a well-conditioned subset of the feature tracks is iteratively selected to accelerate the time-consuming bundle adjustment. The outer loop is a camera registration loop, where the a priori camera rotations are estimated via rotation averaging on multiple orthogonal maximum spanning trees (OMSTs) of the view-graph and used as weak supervision for the registration. The calibrated camera poses that agree with the a priori camera rotations are preferentially registered, and after all the consistent cameras have been calibrated, the remaining cameras are incrementally registered. The results of extensive experiments demonstrate that our system can reconstruct both general and ambiguous image datasets, and our system outperforms many state-of-the-art SfM systems in terms of efficiency and robustness.
资助项目	National Key R&D Program of China[2016YFB0502002] ; Natural Science Foundation of China[61703397] ; Natural Science Foundation of China[61632003] ; Henan Science and Technology Innovation Outstanding Youth Program[184100510009] ; Henan University Scientific and Technological Innovation Team Support Program[19IRTSTHN012]
WOS研究方向	Physical Geography ; Geology ; Remote Sensing ; Imaging Science & Photographic Technology
语种	英语
出版者	ELSEVIER
WOS记录号	WOS:000487765800015
资助机构	National Key R&D Program of China ; Natural Science Foundation of China ; Henan Science and Technology Innovation Outstanding Youth Program ; Henan University Scientific and Technological Innovation Team Support Program
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/26076]
专题	机器人视觉团队
通讯作者	Cui, Hainan; Liu, Hongmin
作者单位	1.Chinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing 100190, Peoples R China 2.Univ Chinese Acad Sci, Beijing 100049, Peoples R China 3.Henan Polytech Univ, Jiaozuo 454000, Henan, Peoples R China 4.Univ Sci & Technol Beijing, Sch Automat & Elect Engn, Beijing 100083, Peoples R China
推荐引用方式 GB/T 7714	Cui, Hainan,Shen, Shuhan,Gao, Wei,et al. Efficient and robust large-scale structure-from-motion via track selection and camera prioritization[J]. ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING,2019,156:202-214.
APA	Cui, Hainan,Shen, Shuhan,Gao, Wei,Liu, Hongmin,&Wang, Zhiheng.(2019).Efficient and robust large-scale structure-from-motion via track selection and camera prioritization.ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING,156,202-214.
MLA	Cui, Hainan,et al."Efficient and robust large-scale structure-from-motion via track selection and camera prioritization".ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING 156(2019):202-214.