CASIA OpenIR  > 模式识别国家重点实验室  > 机器人视觉
High-Quality and Memory-Efficient Volumetric Integration of Depth Maps Using Plane Priors
Liu, Yangdong1,2; Gao, Wei1,2; Hu, Zhanyi1,2
2018-08
Conference Name2018 24th International Conference on Pattern Recognition (ICPR)
Conference DateAugust 20-24, 2018
Conference PlaceBeijing, China
Abstract

Volumetric integration method is widely used to fuse depth maps in dense 3D reconstruction systems. High memory footprint is one of its main disadvantages. We introduce a method to de-noise depth maps and save memory usage during volumetric integration of depth maps with the use of plane priors. We develop a new planar region detection method with the use of depth gradients and then de-noise the planar region of depth maps. During volumetric integration we allocate the voxels and integrate depth maps with the use of plane priors as well. Extensive experiments show that our method saves approximately 30% memory footprint and has higher reconstruction quality compared with some of the current state-of-the-art systems. These characteristics enable our method to be used for 3D scanning on mobile devices which have limited memory resources.

Indexed ByEI
Document Type会议论文
Identifierhttp://ir.ia.ac.cn/handle/173211/23667
Collection模式识别国家重点实验室_机器人视觉
Corresponding AuthorGao, Wei
Affiliation1.NLPR, Institute of Automation, Chinese Academy of Sciences
2.University of Chinese Academy of Sciences
First Author AffilicationChinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing 100190, Peoples R China
Corresponding Author AffilicationChinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing 100190, Peoples R China
Recommended Citation
GB/T 7714
Liu, Yangdong,Gao, Wei,Hu, Zhanyi. High-Quality and Memory-Efficient Volumetric Integration of Depth Maps Using Plane Priors[C],2018.
Files in This Item: Download All
File Name/Size DocType Version Access License
08545525.pdf(1256KB)会议论文 开放获取CC BY-NC-SAView Download
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Liu, Yangdong]'s Articles
[Gao, Wei]'s Articles
[Hu, Zhanyi]'s Articles
Baidu academic
Similar articles in Baidu academic
[Liu, Yangdong]'s Articles
[Gao, Wei]'s Articles
[Hu, Zhanyi]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Liu, Yangdong]'s Articles
[Gao, Wei]'s Articles
[Hu, Zhanyi]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: 08545525.pdf
Format: Adobe PDF
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.