View : 157 Download: 0

Robust and efficient object reconstructions from closed loop sequences

Title
Robust and efficient object reconstructions from closed loop sequences
Authors
Han K.M.Rueda A.J.
Ewha Authors
한경민
SCOPUS Author ID
한경민scopus
Issue Date
2021
Journal Title
Machine Vision and Applications
ISSN
0932-8092JCR Link
Citation
Machine Vision and Applications vol. 32, no. 3
Publisher
Springer Science and Business Media Deutschland GmbH
Indexed
SCIE; SCOPUS WOS scopus
Document Type
Article
Abstract
We propose a new hierarchical structure from motion system for close loop sequences. Our system includes a novel approach for clustering cameras into multiple sets, whose camera poses are initially reconstructed separately and later globally registered w.r.t a single coordinate frame. Each of the multiple sets is robustly reconstructed by a novel guarded least median square protocol. Our method is accelerated by reducing the parameter space of bundle adjustment in the local reconstruction optimizations. We also propose a new synthetic dataset that could be useful in 3D object reconstruction problems. Extensive experiments with both synthetic and real data were carried out to validate our method. Our system presented better results than ACTS and GPE in terms of rotation and translation errors in the camera pose estimations, and the accuracy is quite close to COLMAP while our method is much faster. © 2021, The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature.
Show the fulltext
DOI
10.1007/s00138-021-01193-7
Appears in Collections:
연구기관 > 컴퓨터그래픽스·가상현실연구센터 > Journal papers
Files in This Item:
There are no files associated with this item.
Export
RIS (EndNote)
XLS (Excel)
XML


qrcode

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

BROWSE