A Performance Evaluation of Local Features for Image-Based 3D Reconstruction

doi:10.1109/TIP.2019.2909640

	A Performance Evaluation of Local Features for Image-Based 3D Reconstruction
	Fan, Bin1 ; Kong, Qingqun2 ; Wang, Xinchao 3,4; Wang, Zhiheng 5; Xiang, Shiming1 ; Pan, Chunhong1 ; Fua, Pascal 6
刊名	IEEE TRANSACTIONS ON IMAGE PROCESSING
	2019-10-01
卷号	28 期号:10 页码:4774-4789
关键词	Local feature image reconstruction structure from motion (SFM) 3D vision image matching
ISSN号	1057-7149
DOI	10.1109/TIP.2019.2909640
通讯作者	Wang, Zhiheng(wzhenry@eyou.com)
英文摘要	This paper performs a comprehensive and comparative evaluation of the state-of-the-art local features for the task of image-based 3D reconstruction. The evaluated local features cover the recently developed ones by using powerful machine learning techniques and the elaborately designed handcrafted features. To obtain a comprehensive evaluation, we choose to include both float type features and binary ones. Meanwhile, two kinds of datasets have been used in this evaluation. One is a dataset of many different scene types with groundtruth 3D points, containing images of different scenes captured at fixed positions, for quantitative performance evaluation of different local features in the controlled image capturing situation. The other dataset contains Internet scale image sets of several landmarks with a lot of unrelated images, which is used for qualitative performance evaluation of different local features in the free image collection situation. Our experimental results show that binary features are competent to reconstruct scenes from controlled image sequences with only a fraction of processing time compared to using float type features. However, for the case of a large scale image set with many distracting images, float type features show a clear advantage over binary ones. Currently, the most traditional SIFT is very stable with regard to scene types in this specific task and produces very competitive reconstruction results among all the evaluated local features. Meanwhile, although the learned binary features are not as competitive as the handcrafted ones, learning float type features with CNN is promising but still requires much effort in the future.
资助项目	National Natural Science Foundation of China[61573352] ; National Natural Science Foundation of China[61876180] ; Young Elite Scientists Sponsorship Program by CAST[2018QNRC001] ; Henan Science and Technology Innovation Outstanding Youth Program[184100510009] ; Henan University Scientific and Technological Innovation Team Support Program[19IRTSTHN012]
WOS关键词	DESCRIPTORS ; BINARY ; SCALE ; DETECTORS
WOS研究方向	Computer Science ; Engineering
语种	英语
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
WOS记录号	WOS:000480312800005
资助机构	National Natural Science Foundation of China ; Young Elite Scientists Sponsorship Program by CAST ; 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/173211/27550]
专题	中国科学院自动化研究所
通讯作者	Wang, Zhiheng
作者单位	1.Chinese Acad Sci, Natl Lab Pattern Recognit, Inst Automat, Beijing 100190, Peoples R China 2.Chinese Acad Sci, Inst Automat, Beijing, Peoples R China 3.Univ Chinese Acad Sci, Beijing 100049, Peoples R China 4.Stevens Inst Technol, Dept Comp Sci, Hoboken, NJ 07030 USA 5.Henan Polytech Univ, Sch Comp Sci & Tech, Jiaozuo 454000, Henan, Peoples R China 6.Ecole Polytech Fed Lausanne, CVLab, CH-1015 Lausanne, Switzerland
推荐引用方式 GB/T 7714	Fan, Bin,Kong, Qingqun,Wang, Xinchao,et al. A Performance Evaluation of Local Features for Image-Based 3D Reconstruction[J]. IEEE TRANSACTIONS ON IMAGE PROCESSING,2019,28(10):4774-4789.
APA	Fan, Bin.,Kong, Qingqun.,Wang, Xinchao.,Wang, Zhiheng.,Xiang, Shiming.,...&Fua, Pascal.(2019).A Performance Evaluation of Local Features for Image-Based 3D Reconstruction.IEEE TRANSACTIONS ON IMAGE PROCESSING,28(10),4774-4789.
MLA	Fan, Bin,et al."A Performance Evaluation of Local Features for Image-Based 3D Reconstruction".IEEE TRANSACTIONS ON IMAGE PROCESSING 28.10(2019):4774-4789.