Improved visual odometry system based on kinect RGB-D sensor

Shen Ho Liu; Chen Chien Hsu; Wei Yen Wang; Mei Yung Chen; Yin Tien Wang

doi:10.1109/ICCE-Berlin.2017.8210581

Improved visual odometry system based on kinect RGB-D sensor

Shen Ho Liu, Chen Chien Hsu, Wei Yen Wang, Mei Yung Chen, Yin Tien Wang

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

3 Citations (Scopus)

Abstract

In conventional visual odometry (VO) systems, perspective-n-point (PnP) method and random sample consensus (RANSAC) algorithm are generally used to estimate camera poses. However, heavy computational burden is incurred, and the pose estimations are not reliable as well. Therefore, in this paper, an improved VO system is proposed, where an off-line camera calibration method is used to obtain lesser measurement errors of image features. Moreover, an improved approach of P3P algorithm is proposed to increase the efficiency of the VO system. To validate the performances of the proposed approach, several experiments are conducted based on a Kinect sensor, where accuracy of pose estimations and runtime efficiency are both improved in comparison to the conventional VO algorithms.

Original language	English
Title of host publication	2017 IEEE 7th International Conference on Consumer Electronics - Berlin, ICCE-Berlin 2017
Publisher	IEEE Computer Society
Pages	29-30
Number of pages	2
ISBN (Electronic)	9781509040148
DOIs	https://doi.org/10.1109/ICCE-Berlin.2017.8210581
Publication status	Published - 2017 Dec 14
Event	7th IEEE International Conference on Consumer Electronics - Berlin, ICCE-Berlin 2017 - Berlin, Germany Duration: 2017 Sept 3 → 2017 Sept 6

Publication series

Name	IEEE International Conference on Consumer Electronics - Berlin, ICCE-Berlin
Volume	2017-September
ISSN (Print)	2166-6814
ISSN (Electronic)	2166-6822

Other

Other	7th IEEE International Conference on Consumer Electronics - Berlin, ICCE-Berlin 2017
Country/Territory	Germany
City	Berlin
Period	2017/09/03 → 2017/09/06

Keywords

Perspective-3-point
RANSAC
Visual odometry

ASJC Scopus subject areas

Electrical and Electronic Engineering
Industrial and Manufacturing Engineering
Media Technology

Access to Document

10.1109/ICCE-Berlin.2017.8210581

Cite this

Liu, S. H., Hsu, C. C., Wang, W. Y., Chen, M. Y., & Wang, Y. T. (2017). Improved visual odometry system based on kinect RGB-D sensor. In 2017 IEEE 7th International Conference on Consumer Electronics - Berlin, ICCE-Berlin 2017 (pp. 29-30). (IEEE International Conference on Consumer Electronics - Berlin, ICCE-Berlin; Vol. 2017-September). IEEE Computer Society. https://doi.org/10.1109/ICCE-Berlin.2017.8210581

Improved visual odometry system based on kinect RGB-D sensor. / Liu, Shen Ho; Hsu, Chen Chien ; Wang, Wei Yen et al.
2017 IEEE 7th International Conference on Consumer Electronics - Berlin, ICCE-Berlin 2017. IEEE Computer Society, 2017. p. 29-30 (IEEE International Conference on Consumer Electronics - Berlin, ICCE-Berlin; Vol. 2017-September).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Liu, SH, Hsu, CC , Wang, WY , Chen, MY & Wang, YT 2017, Improved visual odometry system based on kinect RGB-D sensor. in 2017 IEEE 7th International Conference on Consumer Electronics - Berlin, ICCE-Berlin 2017. IEEE International Conference on Consumer Electronics - Berlin, ICCE-Berlin, vol. 2017-September, IEEE Computer Society, pp. 29-30, 7th IEEE International Conference on Consumer Electronics - Berlin, ICCE-Berlin 2017, Berlin, Germany, 2017/09/03. https://doi.org/10.1109/ICCE-Berlin.2017.8210581

@inproceedings{378524cc895541bdb5c80a6f76a03f96,

title = "Improved visual odometry system based on kinect RGB-D sensor",

abstract = "In conventional visual odometry (VO) systems, perspective-n-point (PnP) method and random sample consensus (RANSAC) algorithm are generally used to estimate camera poses. However, heavy computational burden is incurred, and the pose estimations are not reliable as well. Therefore, in this paper, an improved VO system is proposed, where an off-line camera calibration method is used to obtain lesser measurement errors of image features. Moreover, an improved approach of P3P algorithm is proposed to increase the efficiency of the VO system. To validate the performances of the proposed approach, several experiments are conducted based on a Kinect sensor, where accuracy of pose estimations and runtime efficiency are both improved in comparison to the conventional VO algorithms.",

keywords = "Perspective-3-point, RANSAC, Visual odometry",

author = "Liu, {Shen Ho} and Hsu, {Chen Chien} and Wang, {Wei Yen} and Chen, {Mei Yung} and Wang, {Yin Tien}",

note = "Publisher Copyright: {\textcopyright} 2017 IEEE.; 7th IEEE International Conference on Consumer Electronics - Berlin, ICCE-Berlin 2017 ; Conference date: 03-09-2017 Through 06-09-2017",

year = "2017",

month = dec,

day = "14",

doi = "10.1109/ICCE-Berlin.2017.8210581",

language = "English",

series = "IEEE International Conference on Consumer Electronics - Berlin, ICCE-Berlin",

publisher = "IEEE Computer Society",

pages = "29--30",

booktitle = "2017 IEEE 7th International Conference on Consumer Electronics - Berlin, ICCE-Berlin 2017",

}

TY - GEN

T1 - Improved visual odometry system based on kinect RGB-D sensor

AU - Liu, Shen Ho

AU - Hsu, Chen Chien

AU - Wang, Wei Yen

AU - Chen, Mei Yung

AU - Wang, Yin Tien

PY - 2017/12/14

Y1 - 2017/12/14

N2 - In conventional visual odometry (VO) systems, perspective-n-point (PnP) method and random sample consensus (RANSAC) algorithm are generally used to estimate camera poses. However, heavy computational burden is incurred, and the pose estimations are not reliable as well. Therefore, in this paper, an improved VO system is proposed, where an off-line camera calibration method is used to obtain lesser measurement errors of image features. Moreover, an improved approach of P3P algorithm is proposed to increase the efficiency of the VO system. To validate the performances of the proposed approach, several experiments are conducted based on a Kinect sensor, where accuracy of pose estimations and runtime efficiency are both improved in comparison to the conventional VO algorithms.

AB - In conventional visual odometry (VO) systems, perspective-n-point (PnP) method and random sample consensus (RANSAC) algorithm are generally used to estimate camera poses. However, heavy computational burden is incurred, and the pose estimations are not reliable as well. Therefore, in this paper, an improved VO system is proposed, where an off-line camera calibration method is used to obtain lesser measurement errors of image features. Moreover, an improved approach of P3P algorithm is proposed to increase the efficiency of the VO system. To validate the performances of the proposed approach, several experiments are conducted based on a Kinect sensor, where accuracy of pose estimations and runtime efficiency are both improved in comparison to the conventional VO algorithms.

KW - Perspective-3-point

KW - RANSAC

KW - Visual odometry

UR - http://www.scopus.com/inward/record.url?scp=85043977642&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85043977642&partnerID=8YFLogxK

U2 - 10.1109/ICCE-Berlin.2017.8210581

DO - 10.1109/ICCE-Berlin.2017.8210581

M3 - Conference contribution

AN - SCOPUS:85043977642

T3 - IEEE International Conference on Consumer Electronics - Berlin, ICCE-Berlin

SP - 29

EP - 30

BT - 2017 IEEE 7th International Conference on Consumer Electronics - Berlin, ICCE-Berlin 2017

PB - IEEE Computer Society

T2 - 7th IEEE International Conference on Consumer Electronics - Berlin, ICCE-Berlin 2017

Y2 - 3 September 2017 through 6 September 2017

ER -

Improved visual odometry system based on kinect RGB-D sensor

Abstract

Publication series

Other

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this