SLAMBench 3.0: Systematic Automated Reproducible Evaluation of SLAM Systems for Robot Vision Challenges and Scene Understanding

Mihai Bujanca; Paul Gafton; Sajad  Saeedi; Andy Nisbet; Bruno Bodin; Michael F O'Boyle; Andrew J Davison; Paul H Kelly; Graham Riley; Barry Lennox; Mikel Luján; Steve Furber

doi:10.1109/ICRA.2019.8794369

SLAMBench 3.0: Systematic Automated Reproducible Evaluation of SLAM Systems for Robot Vision Challenges and Scene Understanding

Mihai Bujanca, Paul Gafton, Sajad Saeedi, Andy Nisbet, Bruno Bodin, Michael F O'Boyle, Andrew J Davison, Paul H Kelly, Graham Riley, Barry Lennox, Mikel Luján, Steve Furber

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

926 Downloads (Pure)

Abstract

As the SLAM research area matures and the number of SLAM systems available increases, the need for frameworks that can objectively evaluate them against prior work grows. This new version of SLAMBench moves beyond traditional visual SLAM, and provides new support for scene understanding and non-rigid environments (dynamic SLAM). More concretely for dynamic SLAM, SLAMBench 3.0 includes the first publicly available implementation of DynamicFusion, along with an evaluation infrastructure. In addition, we include two SLAM systems (one dense, one sparse) augmented with convolutional neural networks for scene understanding, together with datasets and appropriate metrics. Through a series of use-cases, we demonstrate the newly incorporated algorithms, visulation aids and metrics (6 new metrics, 4 new datasets and 5 new algorithms).

Original language	English
Title of host publication	2019 International Conference on Robotics and Automation (ICRA)
Publisher	IEEE
DOIs	https://doi.org/10.1109/ICRA.2019.8794369
Publication status	Published - 12 Aug 2019

Publication series

Name	2019 International Conference on Robotics and Automation (ICRA)
Publisher	IEEE
ISSN (Print)	1050-4729
ISSN (Electronic)	2577-087X

Access to Document

10.1109/ICRA.2019.8794369Licence: Unspecified

SLAMBench 3.0Final published version, 161 KBLicence: Unspecified

Robotics and Artificial Intelligence for Nuclear (RAIN)
Lennox, B. (PI), Arvin, F. (CoI), Brown, G. (CoI), Carrasco Gomez, J. (CoI), Da Via, C. (CoI), Furber, S. (CoI), Luján, M. (CoI), Watson, S. (CoI), Watts, S. (CoI) & Weightman, A. (CoI)
2/10/17 → 31/03/22
Project: Research

Cite this

Bujanca, M., Gafton, P., Saeedi, S., Nisbet, A., Bodin, B., O'Boyle, M. F., Davison, A. J., Kelly, P. H., Riley, G., Lennox, B., Luján, M., & Furber, S. (2019). SLAMBench 3.0: Systematic Automated Reproducible Evaluation of SLAM Systems for Robot Vision Challenges and Scene Understanding. In 2019 International Conference on Robotics and Automation (ICRA) (2019 International Conference on Robotics and Automation (ICRA)). IEEE. https://doi.org/10.1109/ICRA.2019.8794369

@inproceedings{6f08ab983fe144838d08232f27fa55c1,

title = "SLAMBench 3.0: Systematic Automated Reproducible Evaluation of SLAM Systems for Robot Vision Challenges and Scene Understanding",

abstract = "As the SLAM research area matures and the number of SLAM systems available increases, the need for frameworks that can objectively evaluate them against prior work grows. This new version of SLAMBench moves beyond traditional visual SLAM, and provides new support for scene understanding and non-rigid environments (dynamic SLAM). More concretely for dynamic SLAM, SLAMBench 3.0 includes the first publicly available implementation of DynamicFusion, along with an evaluation infrastructure. In addition, we include two SLAM systems (one dense, one sparse) augmented with convolutional neural networks for scene understanding, together with datasets and appropriate metrics. Through a series of use-cases, we demonstrate the newly incorporated algorithms, visulation aids and metrics (6 new metrics, 4 new datasets and 5 new algorithms).",

author = "Mihai Bujanca and Paul Gafton and Sajad Saeedi and Andy Nisbet and Bruno Bodin and O'Boyle, {Michael F} and Davison, {Andrew J} and Kelly, {Paul H} and Graham Riley and Barry Lennox and Mikel Luj{\'a}n and Steve Furber",

year = "2019",

month = aug,

day = "12",

doi = "10.1109/ICRA.2019.8794369",

language = "English",

series = "2019 International Conference on Robotics and Automation (ICRA)",

publisher = "IEEE",

booktitle = "2019 International Conference on Robotics and Automation (ICRA)",

address = "United States",

}

Bujanca, M, Gafton, P, Saeedi, S, Nisbet, A, Bodin, B, O'Boyle, MF, Davison, AJ, Kelly, PH, Riley, G , Lennox, B, Luján, M & Furber, S 2019, SLAMBench 3.0: Systematic Automated Reproducible Evaluation of SLAM Systems for Robot Vision Challenges and Scene Understanding. in 2019 International Conference on Robotics and Automation (ICRA). 2019 International Conference on Robotics and Automation (ICRA), IEEE. https://doi.org/10.1109/ICRA.2019.8794369

SLAMBench 3.0: Systematic Automated Reproducible Evaluation of SLAM Systems for Robot Vision Challenges and Scene Understanding. / Bujanca, Mihai; Gafton, Paul; Saeedi, Sajad et al.
2019 International Conference on Robotics and Automation (ICRA). IEEE, 2019. (2019 International Conference on Robotics and Automation (ICRA)).

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

TY - GEN

T1 - SLAMBench 3.0

T2 - Systematic Automated Reproducible Evaluation of SLAM Systems for Robot Vision Challenges and Scene Understanding

AU - Bujanca, Mihai

AU - Gafton, Paul

AU - Saeedi, Sajad

AU - Nisbet, Andy

AU - Bodin, Bruno

AU - O'Boyle, Michael F

AU - Davison, Andrew J

AU - Kelly, Paul H

AU - Riley, Graham

AU - Lennox, Barry

AU - Luján, Mikel

AU - Furber, Steve

PY - 2019/8/12

Y1 - 2019/8/12

N2 - As the SLAM research area matures and the number of SLAM systems available increases, the need for frameworks that can objectively evaluate them against prior work grows. This new version of SLAMBench moves beyond traditional visual SLAM, and provides new support for scene understanding and non-rigid environments (dynamic SLAM). More concretely for dynamic SLAM, SLAMBench 3.0 includes the first publicly available implementation of DynamicFusion, along with an evaluation infrastructure. In addition, we include two SLAM systems (one dense, one sparse) augmented with convolutional neural networks for scene understanding, together with datasets and appropriate metrics. Through a series of use-cases, we demonstrate the newly incorporated algorithms, visulation aids and metrics (6 new metrics, 4 new datasets and 5 new algorithms).

AB - As the SLAM research area matures and the number of SLAM systems available increases, the need for frameworks that can objectively evaluate them against prior work grows. This new version of SLAMBench moves beyond traditional visual SLAM, and provides new support for scene understanding and non-rigid environments (dynamic SLAM). More concretely for dynamic SLAM, SLAMBench 3.0 includes the first publicly available implementation of DynamicFusion, along with an evaluation infrastructure. In addition, we include two SLAM systems (one dense, one sparse) augmented with convolutional neural networks for scene understanding, together with datasets and appropriate metrics. Through a series of use-cases, we demonstrate the newly incorporated algorithms, visulation aids and metrics (6 new metrics, 4 new datasets and 5 new algorithms).

U2 - 10.1109/ICRA.2019.8794369

DO - 10.1109/ICRA.2019.8794369

M3 - Conference contribution

T3 - 2019 International Conference on Robotics and Automation (ICRA)

BT - 2019 International Conference on Robotics and Automation (ICRA)

PB - IEEE

ER -

Bujanca M, Gafton P, Saeedi S, Nisbet A, Bodin B, O'Boyle MF et al. SLAMBench 3.0: Systematic Automated Reproducible Evaluation of SLAM Systems for Robot Vision Challenges and Scene Understanding. In 2019 International Conference on Robotics and Automation (ICRA). IEEE. 2019. (2019 International Conference on Robotics and Automation (ICRA)). doi: 10.1109/ICRA.2019.8794369

SLAMBench 3.0: Systematic Automated Reproducible Evaluation of SLAM Systems for Robot Vision Challenges and Scene Understanding

Abstract

Publication series

Access to Document

Fingerprint

Projects

Robotics and Artificial Intelligence for Nuclear (RAIN)

Cite this