3D Surfel Map-Aided Visual Relocalization with Learned Descriptors

Haoyang Ye; Huaiyang Huang; Marco Hutter; Timothy Sandy; Ming Liu

doi:10.1109/ICRA48506.2021.9561005

3D Surfel Map-Aided Visual Relocalization with Learned Descriptors

Haoyang Ye
, Huaiyang Huang
, Marco Hutter
, Timothy Sandy
, Ming Liu

Department of Computer Science and Engineering

Research output: Chapter in Book/Conference Proceeding/Report › Conference Paper published in a book › peer-review

4 Citations (Scopus)

Abstract

In this paper, we introduce a method for visual relocalization using the geometric information from a 3D surfel map. A visual database is first built by global indices from the 3D surfel map rendering, which provides associations between image points and 3D surfels. Surfel reprojection constraints are utilized to optimize the keyframe poses and map points in the visual database. A hierarchical camera relocalization algorithm then utilizes the visual database to estimate 6-DoF camera poses. Learned descriptors are further used to improve the performance in challenging cases. We present evaluation under real-world conditions and simulation to show the effectiveness and efficiency of our method, and make the final camera poses consistently well aligned with the 3D environment.

Original language	English
Title of host publication	2021 IEEE International Conference on Robotics and Automation, ICRA 2021
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	5574-5581
Number of pages	8
ISBN (Electronic)	9781728190778
DOIs	https://doi.org/10.1109/ICRA48506.2021.9561005
Publication status	Published - 2021
Event	2021 IEEE International Conference on Robotics and Automation, ICRA 2021 - Xi'an, China Duration: 30 May 2021 → 5 Jun 2021

Publication series

Name	Proceedings - IEEE International Conference on Robotics and Automation
Volume	2021-May
ISSN (Print)	1050-4729

Conference

Conference	2021 IEEE International Conference on Robotics and Automation, ICRA 2021
Country/Territory	China
City	Xi'an
Period	30/05/21 → 5/06/21

Bibliographical note

Publisher Copyright:
© 2021 IEEE

Access to Document

10.1109/ICRA48506.2021.9561005

Cite this

Ye, H., Huang, H., Hutter, M., Sandy, T., & Liu, M. (2021). 3D Surfel Map-Aided Visual Relocalization with Learned Descriptors. In 2021 IEEE International Conference on Robotics and Automation, ICRA 2021 (pp. 5574-5581). (Proceedings - IEEE International Conference on Robotics and Automation; Vol. 2021-May). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICRA48506.2021.9561005

@inproceedings{1f13cfc4084b4aef80939f2f1e8080c9,

title = "3D Surfel Map-Aided Visual Relocalization with Learned Descriptors",

abstract = "In this paper, we introduce a method for visual relocalization using the geometric information from a 3D surfel map. A visual database is first built by global indices from the 3D surfel map rendering, which provides associations between image points and 3D surfels. Surfel reprojection constraints are utilized to optimize the keyframe poses and map points in the visual database. A hierarchical camera relocalization algorithm then utilizes the visual database to estimate 6-DoF camera poses. Learned descriptors are further used to improve the performance in challenging cases. We present evaluation under real-world conditions and simulation to show the effectiveness and efficiency of our method, and make the final camera poses consistently well aligned with the 3D environment.",

author = "Haoyang Ye and Huaiyang Huang and Marco Hutter and Timothy Sandy and Ming Liu",

note = "Publisher Copyright: {\textcopyright} 2021 IEEE; 2021 IEEE International Conference on Robotics and Automation, ICRA 2021 ; Conference date: 30-05-2021 Through 05-06-2021",

year = "2021",

doi = "10.1109/ICRA48506.2021.9561005",

language = "English",

series = "Proceedings - IEEE International Conference on Robotics and Automation",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "5574--5581",

booktitle = "2021 IEEE International Conference on Robotics and Automation, ICRA 2021",

}

Ye, H, Huang, H, Hutter, M, Sandy, T & Liu, M 2021, 3D Surfel Map-Aided Visual Relocalization with Learned Descriptors. in 2021 IEEE International Conference on Robotics and Automation, ICRA 2021. Proceedings - IEEE International Conference on Robotics and Automation, vol. 2021-May, Institute of Electrical and Electronics Engineers Inc., pp. 5574-5581, 2021 IEEE International Conference on Robotics and Automation, ICRA 2021, Xi'an, China, 30/05/21. https://doi.org/10.1109/ICRA48506.2021.9561005

3D Surfel Map-Aided Visual Relocalization with Learned Descriptors. / Ye, Haoyang; Huang, Huaiyang; Hutter, Marco et al.
2021 IEEE International Conference on Robotics and Automation, ICRA 2021. Institute of Electrical and Electronics Engineers Inc., 2021. p. 5574-5581 (Proceedings - IEEE International Conference on Robotics and Automation; Vol. 2021-May).

Research output: Chapter in Book/Conference Proceeding/Report › Conference Paper published in a book › peer-review

TY - GEN

T1 - 3D Surfel Map-Aided Visual Relocalization with Learned Descriptors

AU - Ye, Haoyang

AU - Huang, Huaiyang

AU - Hutter, Marco

AU - Sandy, Timothy

AU - Liu, Ming

PY - 2021

Y1 - 2021

N2 - In this paper, we introduce a method for visual relocalization using the geometric information from a 3D surfel map. A visual database is first built by global indices from the 3D surfel map rendering, which provides associations between image points and 3D surfels. Surfel reprojection constraints are utilized to optimize the keyframe poses and map points in the visual database. A hierarchical camera relocalization algorithm then utilizes the visual database to estimate 6-DoF camera poses. Learned descriptors are further used to improve the performance in challenging cases. We present evaluation under real-world conditions and simulation to show the effectiveness and efficiency of our method, and make the final camera poses consistently well aligned with the 3D environment.

AB - In this paper, we introduce a method for visual relocalization using the geometric information from a 3D surfel map. A visual database is first built by global indices from the 3D surfel map rendering, which provides associations between image points and 3D surfels. Surfel reprojection constraints are utilized to optimize the keyframe poses and map points in the visual database. A hierarchical camera relocalization algorithm then utilizes the visual database to estimate 6-DoF camera poses. Learned descriptors are further used to improve the performance in challenging cases. We present evaluation under real-world conditions and simulation to show the effectiveness and efficiency of our method, and make the final camera poses consistently well aligned with the 3D environment.

UR - https://www.scopus.com/pages/publications/85125443643

U2 - 10.1109/ICRA48506.2021.9561005

DO - 10.1109/ICRA48506.2021.9561005

M3 - Conference Paper published in a book

AN - SCOPUS:85125443643

T3 - Proceedings - IEEE International Conference on Robotics and Automation

SP - 5574

EP - 5581

BT - 2021 IEEE International Conference on Robotics and Automation, ICRA 2021

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2021 IEEE International Conference on Robotics and Automation, ICRA 2021

Y2 - 30 May 2021 through 5 June 2021

ER -

Ye H, Huang H, Hutter M, Sandy T, Liu M. 3D Surfel Map-Aided Visual Relocalization with Learned Descriptors. In 2021 IEEE International Conference on Robotics and Automation, ICRA 2021. Institute of Electrical and Electronics Engineers Inc. 2021. p. 5574-5581. (Proceedings - IEEE International Conference on Robotics and Automation). doi: 10.1109/ICRA48506.2021.9561005

3D Surfel Map-Aided Visual Relocalization with Learned Descriptors

Abstract

Publication series

Conference

Bibliographical note

Access to Document

Other files and links

Fingerprint

Cite this