Effective piecewise planar modeling based on sparse 3D points and convolutional neural network

doi:10.1016/j.neucom.2019.10.026

	Effective piecewise planar modeling based on sparse 3D points and convolutional neural network
	Wang, Wei 1; Gao, Wei2 ; Hu, Zhanyi2
发表期刊	NEUROCOMPUTING
ISSN	0925-2312
	2020-02-22
卷号	378 页码:350-363
通讯作者	Wang, Wei(wangwei@zknu.cn)
摘要	Piecewise planar stereo methods can approximately reconstruct the complete structures of a scene by overcoming challenging difficulties (e.g., poorly textured regions) that pixel-level stereo methods cannot resolve. In this paper, a novel plane assignment cost is first constructed by incorporating scene structure priors and high-level image features obtained by convolutional neural network (CNN). Then, the piecewise planar scene structures are reconstructed in a progressive manner that jointly optimizes image regions (or superpixels) and their associated planes, followed by a global plane assignment optimization under a Markov Random Field (MRF) framework. Experimental results on a variety of urban scenes confirm that the proposed method can effectively reconstruct the complete structures of a scene from only sparse three-dimensional (3D) points with high efficiency and accuracy and can achieve superior results compared with state-of-the-art methods. (C) 2019 Elsevier B.V. All rights reserved.
关键词	Urban scene Piecewise planar stereo Markov Random Field Image over-segmentation Convolutional neural network
DOI	10.1016/j.neucom.2019.10.026
关键词[WOS]	SUPERPIXELS
收录类别	SCI
语种	英语
资助项目	National Key R&D Program of China[2016YFB0502002] ; National Natural Science Foundation of China[U1805264] ; National Natural Science Foundation of China[617724 44] ; National Natural Science Foundation of China[61421004] ; National Natural Science Foundation of China[61873264] ; Natural Science Foundation of Henan Province[162300410347] ; Key Scientific and Technological Project of Henan Province[162102310589] ; Key Scientific and Technological Project of Henan Province[192102210279] ; National Key R&D Program of China[2016YFB0502002] ; National Natural Science Foundation of China[U1805264] ; National Natural Science Foundation of China[617724 44] ; National Natural Science Foundation of China[61421004] ; National Natural Science Foundation of China[61873264] ; Natural Science Foundation of Henan Province[162300410347] ; Key Scientific and Technological Project of Henan Province[162102310589] ; Key Scientific and Technological Project of Henan Province[192102210279]
项目资助者	National Key R&D Program of China ; National Natural Science Foundation of China ; Natural Science Foundation of Henan Province ; Key Scientific and Technological Project of Henan Province
WOS研究方向	Computer Science
WOS类目	Computer Science, Artificial Intelligence
WOS记录号	WOS:000506202200031
出版者	ELSEVIER
引用统计
文献类型	期刊论文
条目标识符	http://ir.ia.ac.cn/handle/173211/29448
专题	多模态人工智能系统全国重点实验室_机器人视觉
通讯作者	Wang, Wei
作者单位	1.Zhoukou Normal Univ, Sch Network Engn, Zhoukou 466001, Peoples R China 2.Chinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing 100190, Peoples R China
推荐引用方式 GB/T 7714	Wang, Wei,Gao, Wei,Hu, Zhanyi. Effective piecewise planar modeling based on sparse 3D points and convolutional neural network[J]. NEUROCOMPUTING,2020,378:350-363.
APA	Wang, Wei,Gao, Wei,&Hu, Zhanyi.(2020).Effective piecewise planar modeling based on sparse 3D points and convolutional neural network.NEUROCOMPUTING,378,350-363.
MLA	Wang, Wei,et al."Effective piecewise planar modeling based on sparse 3D points and convolutional neural network".NEUROCOMPUTING 378(2020):350-363.