Mo-SLAM: Multi object SLAM with run-time object discovery through duplicates

Thanuja Dharmasiri; Vincent Lui; Tom Drummond

doi:10.1109/IROS.2016.7759203

Mo-SLAM: Multi object SLAM with run-time object discovery through duplicates

Thanuja Dharmasiri, Vincent Lui, Tom Drummond

Electrical and Computer Systems Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research › peer-review

14 Citations (Scopus)

Abstract

In this paper, we present MO-SLAM, a novel visual SLAM system that is capable of detecting duplicate objects in the scene during run-time without requiring an offline training stage to pre-populate a database of objects. Instead, we propose a novel method to detect landmarks that belong to duplicate objects. Further, we show how landmarks belonging to duplicate objects can be converted to first-order entities which generate additional constraints for optimizing the map. We evaluate the performance of MO-SLAM with extensive experiments on both synthetic and real data, where the experimental results verify the capabilities of MO-SLAM in detecting duplicate objects and using these constraints to improve the accuracy of the map.

Original language	English
Title of host publication	2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2016)
Subtitle of host publication	Daejeon, South Korea, 9-14 October, 2016
Editors	Wolfram Burgard
Place of Publication	Piscataway, NJ
Publisher	IEEE, Institute of Electrical and Electronics Engineers
Pages	1214-1221
Number of pages	8
ISBN (Electronic)	9781509037629
ISBN (Print)	9781509037636
DOIs	https://doi.org/10.1109/IROS.2016.7759203
Publication status	Published - 28 Nov 2016
Event	IEEE/RSJ International Conference on Intelligent Robots and Systems 2016 - Daejeon, Korea, Republic of (South) Duration: 9 Oct 2016 → 14 Oct 2016 http://www.iros2016.org/ https://ieeexplore.ieee.org/xpl/conhome/7743711/proceeding (Proceedings)

Conference

Conference	IEEE/RSJ International Conference on Intelligent Robots and Systems 2016
Abbreviated title	IROS 2016
Country/Territory	Korea, Republic of (South)
City	Daejeon
Period	9/10/16 → 14/10/16
Internet address	http://www.iros2016.org/ https://ieeexplore.ieee.org/xpl/conhome/7743711/proceeding (Proceedings)

Access to Document

10.1109/IROS.2016.7759203

Cite this

Dharmasiri, T., Lui, V., & Drummond, T. (2016). Mo-SLAM: Multi object SLAM with run-time object discovery through duplicates. In W. Burgard (Ed.), 2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2016) : Daejeon, South Korea, 9-14 October, 2016 (pp. 1214-1221). [7759203] IEEE, Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/IROS.2016.7759203

Dharmasiri, Thanuja ; Lui, Vincent ; Drummond, Tom. / Mo-SLAM : Multi object SLAM with run-time object discovery through duplicates. 2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2016) : Daejeon, South Korea, 9-14 October, 2016. editor / Wolfram Burgard. Piscataway, NJ : IEEE, Institute of Electrical and Electronics Engineers, 2016. pp. 1214-1221

@inproceedings{c8b61fb928484865a2c1c927ed450b9f,

title = "Mo-SLAM: Multi object SLAM with run-time object discovery through duplicates",

abstract = "In this paper, we present MO-SLAM, a novel visual SLAM system that is capable of detecting duplicate objects in the scene during run-time without requiring an offline training stage to pre-populate a database of objects. Instead, we propose a novel method to detect landmarks that belong to duplicate objects. Further, we show how landmarks belonging to duplicate objects can be converted to first-order entities which generate additional constraints for optimizing the map. We evaluate the performance of MO-SLAM with extensive experiments on both synthetic and real data, where the experimental results verify the capabilities of MO-SLAM in detecting duplicate objects and using these constraints to improve the accuracy of the map.",

author = "Thanuja Dharmasiri and Vincent Lui and Tom Drummond",

year = "2016",

month = nov,

day = "28",

doi = "10.1109/IROS.2016.7759203",

language = "English",

isbn = "9781509037636 ",

pages = "1214--1221",

editor = "Burgard, {Wolfram }",

booktitle = "2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2016)",

publisher = "IEEE, Institute of Electrical and Electronics Engineers",

address = "United States of America",

note = "IEEE/RSJ International Conference on Intelligent Robots and Systems 2016 , IROS 2016 ; Conference date: 09-10-2016 Through 14-10-2016",

url = "http://www.iros2016.org/, https://ieeexplore.ieee.org/xpl/conhome/7743711/proceeding",

}

Dharmasiri, T, Lui, V & Drummond, T 2016, Mo-SLAM: Multi object SLAM with run-time object discovery through duplicates. in W Burgard (ed.), 2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2016) : Daejeon, South Korea, 9-14 October, 2016., 7759203, IEEE, Institute of Electrical and Electronics Engineers, Piscataway, NJ , pp. 1214-1221, IEEE/RSJ International Conference on Intelligent Robots and Systems 2016 , Daejeon, Korea, Republic of (South), 9/10/16. https://doi.org/10.1109/IROS.2016.7759203

Mo-SLAM : Multi object SLAM with run-time object discovery through duplicates. / Dharmasiri, Thanuja; Lui, Vincent; Drummond, Tom.

2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2016) : Daejeon, South Korea, 9-14 October, 2016. ed. / Wolfram Burgard. Piscataway, NJ : IEEE, Institute of Electrical and Electronics Engineers, 2016. p. 1214-1221 7759203.

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research › peer-review

TY - GEN

T1 - Mo-SLAM

T2 - IEEE/RSJ International Conference on Intelligent Robots and Systems 2016

AU - Dharmasiri, Thanuja

AU - Lui, Vincent

AU - Drummond, Tom

PY - 2016/11/28

Y1 - 2016/11/28

N2 - In this paper, we present MO-SLAM, a novel visual SLAM system that is capable of detecting duplicate objects in the scene during run-time without requiring an offline training stage to pre-populate a database of objects. Instead, we propose a novel method to detect landmarks that belong to duplicate objects. Further, we show how landmarks belonging to duplicate objects can be converted to first-order entities which generate additional constraints for optimizing the map. We evaluate the performance of MO-SLAM with extensive experiments on both synthetic and real data, where the experimental results verify the capabilities of MO-SLAM in detecting duplicate objects and using these constraints to improve the accuracy of the map.

AB - In this paper, we present MO-SLAM, a novel visual SLAM system that is capable of detecting duplicate objects in the scene during run-time without requiring an offline training stage to pre-populate a database of objects. Instead, we propose a novel method to detect landmarks that belong to duplicate objects. Further, we show how landmarks belonging to duplicate objects can be converted to first-order entities which generate additional constraints for optimizing the map. We evaluate the performance of MO-SLAM with extensive experiments on both synthetic and real data, where the experimental results verify the capabilities of MO-SLAM in detecting duplicate objects and using these constraints to improve the accuracy of the map.

UR - http://www.scopus.com/inward/record.url?scp=85006355346&partnerID=8YFLogxK

U2 - 10.1109/IROS.2016.7759203

DO - 10.1109/IROS.2016.7759203

M3 - Conference Paper

AN - SCOPUS:85006355346

SN - 9781509037636

SP - 1214

EP - 1221

BT - 2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2016)

A2 - Burgard, Wolfram

PB - IEEE, Institute of Electrical and Electronics Engineers

CY - Piscataway, NJ

Y2 - 9 October 2016 through 14 October 2016

ER -

Dharmasiri T, Lui V, Drummond T. Mo-SLAM: Multi object SLAM with run-time object discovery through duplicates. In Burgard W, editor, 2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2016) : Daejeon, South Korea, 9-14 October, 2016. Piscataway, NJ : IEEE, Institute of Electrical and Electronics Engineers. 2016. p. 1214-1221. 7759203 https://doi.org/10.1109/IROS.2016.7759203

Mo-SLAM: Multi object SLAM with run-time object discovery through duplicates

Abstract

Conference

Access to Document

Other files and links

Cite this