Line-based SLAM considering prior distribution of distance and angle of line features in an urban environment

Kei Uehara, Hideo Saito, Kosuke Hara

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In this paper, we propose a line-based SLAM from an image sequence captured by a camera mounted on a vehicle in consideration with the prior distribution of line features that detected in an urban environments. Since such scenes captured by the vehicle in urban envirounments can be expected to include a lot of line segments detected from road markings and buildings, we employ line segments as features for our SLAM. We use additional prior regarding the line segments so that we can improve the accuracy of the SLAM. We assume that the angle of the vector of the line segments to the vehicle’s direction of travel conform to four-component Gaussian mixture distribution. We define a new cost function considering the prior distribution and optimize the relative camera pose, position, and the 3D line segments by bundle adjustment. The prior distribution is also extended into 2D, the distance and angle of the line segments. In addition, we make digital maps from the detected line segments. Our method increases the accuracy of localization and corrects tilted lines in the digital maps. We apply our method to both the single-camera system and the multi-camera system for demonstrate the accuracy improvement by the prior distribution of distance and angle of line features.

Original languageEnglish
Title of host publicationComputer Vision, Imaging and Computer Graphics – Theory and Applications - 12th International Joint Conference, VISIGRAPP 2017, Revised Selected Papers
EditorsPaul Richard, Alexandru Telea, Takehiko Yamaguchi, Alain Tremeau, Dominique Bechmann, Ana Paula Cláudio, Lars Linsen, Francisco Imai
PublisherSpringer Verlag
Pages105-127
Number of pages23
ISBN (Print)9783030122089
DOIs
Publication statusPublished - 2019 Jan 1
Event12th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications, VISIGRAPP 2017 - Porto, Portugal
Duration: 2017 Feb 272017 Mar 1

Publication series

NameCommunications in Computer and Information Science
Volume983
ISSN (Print)1865-0929

Other

Other12th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications, VISIGRAPP 2017
CountryPortugal
CityPorto
Period17/2/2717/3/1

Fingerprint

Simultaneous Localization and Mapping
Line segment
Prior distribution
Cameras
Angle
Line
Camera
Cost functions
Mixture Distribution
Gaussian Mixture
Image Sequence
Cost Function
Gaussian distribution
Bundle
Adjustment
Optimise
Demonstrate

Keywords

  • Gaussian distribution
  • Line-based SLAM
  • Manhattan world assumption
  • Road markings

ASJC Scopus subject areas

  • Computer Science(all)
  • Mathematics(all)

Cite this

Uehara, K., Saito, H., & Hara, K. (2019). Line-based SLAM considering prior distribution of distance and angle of line features in an urban environment. In P. Richard, A. Telea, T. Yamaguchi, A. Tremeau, D. Bechmann, A. P. Cláudio, L. Linsen, ... F. Imai (Eds.), Computer Vision, Imaging and Computer Graphics – Theory and Applications - 12th International Joint Conference, VISIGRAPP 2017, Revised Selected Papers (pp. 105-127). (Communications in Computer and Information Science; Vol. 983). Springer Verlag. https://doi.org/10.1007/978-3-030-12209-6_6

Line-based SLAM considering prior distribution of distance and angle of line features in an urban environment. / Uehara, Kei; Saito, Hideo; Hara, Kosuke.

Computer Vision, Imaging and Computer Graphics – Theory and Applications - 12th International Joint Conference, VISIGRAPP 2017, Revised Selected Papers. ed. / Paul Richard; Alexandru Telea; Takehiko Yamaguchi; Alain Tremeau; Dominique Bechmann; Ana Paula Cláudio; Lars Linsen; Francisco Imai. Springer Verlag, 2019. p. 105-127 (Communications in Computer and Information Science; Vol. 983).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Uehara, K, Saito, H & Hara, K 2019, Line-based SLAM considering prior distribution of distance and angle of line features in an urban environment. in P Richard, A Telea, T Yamaguchi, A Tremeau, D Bechmann, AP Cláudio, L Linsen & F Imai (eds), Computer Vision, Imaging and Computer Graphics – Theory and Applications - 12th International Joint Conference, VISIGRAPP 2017, Revised Selected Papers. Communications in Computer and Information Science, vol. 983, Springer Verlag, pp. 105-127, 12th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications, VISIGRAPP 2017, Porto, Portugal, 17/2/27. https://doi.org/10.1007/978-3-030-12209-6_6
Uehara K, Saito H, Hara K. Line-based SLAM considering prior distribution of distance and angle of line features in an urban environment. In Richard P, Telea A, Yamaguchi T, Tremeau A, Bechmann D, Cláudio AP, Linsen L, Imai F, editors, Computer Vision, Imaging and Computer Graphics – Theory and Applications - 12th International Joint Conference, VISIGRAPP 2017, Revised Selected Papers. Springer Verlag. 2019. p. 105-127. (Communications in Computer and Information Science). https://doi.org/10.1007/978-3-030-12209-6_6
Uehara, Kei ; Saito, Hideo ; Hara, Kosuke. / Line-based SLAM considering prior distribution of distance and angle of line features in an urban environment. Computer Vision, Imaging and Computer Graphics – Theory and Applications - 12th International Joint Conference, VISIGRAPP 2017, Revised Selected Papers. editor / Paul Richard ; Alexandru Telea ; Takehiko Yamaguchi ; Alain Tremeau ; Dominique Bechmann ; Ana Paula Cláudio ; Lars Linsen ; Francisco Imai. Springer Verlag, 2019. pp. 105-127 (Communications in Computer and Information Science).
@inproceedings{7f81f2cc82d94b37a97e8ef263b9ad58,
title = "Line-based SLAM considering prior distribution of distance and angle of line features in an urban environment",
abstract = "In this paper, we propose a line-based SLAM from an image sequence captured by a camera mounted on a vehicle in consideration with the prior distribution of line features that detected in an urban environments. Since such scenes captured by the vehicle in urban envirounments can be expected to include a lot of line segments detected from road markings and buildings, we employ line segments as features for our SLAM. We use additional prior regarding the line segments so that we can improve the accuracy of the SLAM. We assume that the angle of the vector of the line segments to the vehicle’s direction of travel conform to four-component Gaussian mixture distribution. We define a new cost function considering the prior distribution and optimize the relative camera pose, position, and the 3D line segments by bundle adjustment. The prior distribution is also extended into 2D, the distance and angle of the line segments. In addition, we make digital maps from the detected line segments. Our method increases the accuracy of localization and corrects tilted lines in the digital maps. We apply our method to both the single-camera system and the multi-camera system for demonstrate the accuracy improvement by the prior distribution of distance and angle of line features.",
keywords = "Gaussian distribution, Line-based SLAM, Manhattan world assumption, Road markings",
author = "Kei Uehara and Hideo Saito and Kosuke Hara",
year = "2019",
month = "1",
day = "1",
doi = "10.1007/978-3-030-12209-6_6",
language = "English",
isbn = "9783030122089",
series = "Communications in Computer and Information Science",
publisher = "Springer Verlag",
pages = "105--127",
editor = "Paul Richard and Alexandru Telea and Takehiko Yamaguchi and Alain Tremeau and Dominique Bechmann and Cl{\'a}udio, {Ana Paula} and Lars Linsen and Francisco Imai",
booktitle = "Computer Vision, Imaging and Computer Graphics – Theory and Applications - 12th International Joint Conference, VISIGRAPP 2017, Revised Selected Papers",
address = "Germany",

}

TY - GEN

T1 - Line-based SLAM considering prior distribution of distance and angle of line features in an urban environment

AU - Uehara, Kei

AU - Saito, Hideo

AU - Hara, Kosuke

PY - 2019/1/1

Y1 - 2019/1/1

N2 - In this paper, we propose a line-based SLAM from an image sequence captured by a camera mounted on a vehicle in consideration with the prior distribution of line features that detected in an urban environments. Since such scenes captured by the vehicle in urban envirounments can be expected to include a lot of line segments detected from road markings and buildings, we employ line segments as features for our SLAM. We use additional prior regarding the line segments so that we can improve the accuracy of the SLAM. We assume that the angle of the vector of the line segments to the vehicle’s direction of travel conform to four-component Gaussian mixture distribution. We define a new cost function considering the prior distribution and optimize the relative camera pose, position, and the 3D line segments by bundle adjustment. The prior distribution is also extended into 2D, the distance and angle of the line segments. In addition, we make digital maps from the detected line segments. Our method increases the accuracy of localization and corrects tilted lines in the digital maps. We apply our method to both the single-camera system and the multi-camera system for demonstrate the accuracy improvement by the prior distribution of distance and angle of line features.

AB - In this paper, we propose a line-based SLAM from an image sequence captured by a camera mounted on a vehicle in consideration with the prior distribution of line features that detected in an urban environments. Since such scenes captured by the vehicle in urban envirounments can be expected to include a lot of line segments detected from road markings and buildings, we employ line segments as features for our SLAM. We use additional prior regarding the line segments so that we can improve the accuracy of the SLAM. We assume that the angle of the vector of the line segments to the vehicle’s direction of travel conform to four-component Gaussian mixture distribution. We define a new cost function considering the prior distribution and optimize the relative camera pose, position, and the 3D line segments by bundle adjustment. The prior distribution is also extended into 2D, the distance and angle of the line segments. In addition, we make digital maps from the detected line segments. Our method increases the accuracy of localization and corrects tilted lines in the digital maps. We apply our method to both the single-camera system and the multi-camera system for demonstrate the accuracy improvement by the prior distribution of distance and angle of line features.

KW - Gaussian distribution

KW - Line-based SLAM

KW - Manhattan world assumption

KW - Road markings

UR - http://www.scopus.com/inward/record.url?scp=85061148908&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85061148908&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-12209-6_6

DO - 10.1007/978-3-030-12209-6_6

M3 - Conference contribution

SN - 9783030122089

T3 - Communications in Computer and Information Science

SP - 105

EP - 127

BT - Computer Vision, Imaging and Computer Graphics – Theory and Applications - 12th International Joint Conference, VISIGRAPP 2017, Revised Selected Papers

A2 - Richard, Paul

A2 - Telea, Alexandru

A2 - Yamaguchi, Takehiko

A2 - Tremeau, Alain

A2 - Bechmann, Dominique

A2 - Cláudio, Ana Paula

A2 - Linsen, Lars

A2 - Imai, Francisco

PB - Springer Verlag

ER -