Fusing semantics and motion state detection for robust visual SLAM

Achieving robust pose tracking and mapping in highly dynamic environments is a major challenge faced by existing visual SLAM (vSLAM) systems. In this paper, we increase the robustness of existing vSLAM by accurately removing moving objects from the scene so that they will not contribute to pose esti...

Full description

Saved in:
Bibliographic Details
Main Authors: Singh, Gaurav, Wu, Meiqing, Lam, Siew-Kei
Other Authors: College of Computing and Data Science
Format: Conference or Workshop Item
Language:English
Published: 2024
Subjects:
Online Access:https://hdl.handle.net/10356/178588
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
id sg-ntu-dr.10356-178588
record_format dspace
spelling sg-ntu-dr.10356-1785882024-06-27T01:28:56Z Fusing semantics and motion state detection for robust visual SLAM Singh, Gaurav Wu, Meiqing Lam, Siew-Kei College of Computing and Data Science School of Computer Science and Engineering 2020 IEEE Winter Conference on Applications of Computer Vision (WACV) Computer and Information Science Semantics Robustness Achieving robust pose tracking and mapping in highly dynamic environments is a major challenge faced by existing visual SLAM (vSLAM) systems. In this paper, we increase the robustness of existing vSLAM by accurately removing moving objects from the scene so that they will not contribute to pose estimation and mapping. Specifically, semantic information is fused with motion states of the scene via a probability framework to enable accurate and robust moving object extraction in order to retain the useful features for pose estimation and mapping. Our work highlights the importance of distinguishing between motion states of potential moving objects for vSLAM in highly dynamic environments. The proposed method can be integrated into existing vSLAM systems to increase their robustness in dynamic environments without incurring much computation cost. We provide extensive experimental results on three well-known datasets to show that the proposed technique outperforms existing vSLAM methods in indoor and outdoor environments, under various scenarios such as crowded scenes. 2024-06-27T01:28:56Z 2024-06-27T01:28:56Z 2020 Conference Paper Singh, G., Wu, M. & Lam, S. (2020). Fusing semantics and motion state detection for robust visual SLAM. 2020 IEEE Winter Conference on Applications of Computer Vision (WACV), 2753-2762. https://dx.doi.org/10.1109/WACV45572.2020.9093359 978-1-7281-6553-0 https://hdl.handle.net/10356/178588 10.1109/WACV45572.2020.9093359 2-s2.0-85085519132 2753 2762 en © 2020 IEEE. All rights reserved.
institution Nanyang Technological University
building NTU Library
continent Asia
country Singapore
Singapore
content_provider NTU Library
collection DR-NTU
language English
topic Computer and Information Science
Semantics
Robustness
spellingShingle Computer and Information Science
Semantics
Robustness
Singh, Gaurav
Wu, Meiqing
Lam, Siew-Kei
Fusing semantics and motion state detection for robust visual SLAM
description Achieving robust pose tracking and mapping in highly dynamic environments is a major challenge faced by existing visual SLAM (vSLAM) systems. In this paper, we increase the robustness of existing vSLAM by accurately removing moving objects from the scene so that they will not contribute to pose estimation and mapping. Specifically, semantic information is fused with motion states of the scene via a probability framework to enable accurate and robust moving object extraction in order to retain the useful features for pose estimation and mapping. Our work highlights the importance of distinguishing between motion states of potential moving objects for vSLAM in highly dynamic environments. The proposed method can be integrated into existing vSLAM systems to increase their robustness in dynamic environments without incurring much computation cost. We provide extensive experimental results on three well-known datasets to show that the proposed technique outperforms existing vSLAM methods in indoor and outdoor environments, under various scenarios such as crowded scenes.
author2 College of Computing and Data Science
author_facet College of Computing and Data Science
Singh, Gaurav
Wu, Meiqing
Lam, Siew-Kei
format Conference or Workshop Item
author Singh, Gaurav
Wu, Meiqing
Lam, Siew-Kei
author_sort Singh, Gaurav
title Fusing semantics and motion state detection for robust visual SLAM
title_short Fusing semantics and motion state detection for robust visual SLAM
title_full Fusing semantics and motion state detection for robust visual SLAM
title_fullStr Fusing semantics and motion state detection for robust visual SLAM
title_full_unstemmed Fusing semantics and motion state detection for robust visual SLAM
title_sort fusing semantics and motion state detection for robust visual slam
publishDate 2024
url https://hdl.handle.net/10356/178588
_version_ 1806059756830326784