Sequence matching enhanced 3D place recognition using candidate rearrangement
Abstract Deep‐learning‐based 3D place recognition has received more attention since the data‐driven fashion is widely used for the 3D point cloud applications. Most of the existing deep‐learning‐based 3D place recognition methods only utilise a single scene for place recognition. However, a single s...
Main Authors: | , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Wiley
2022-09-01
|
Series: | IET Cyber-systems and Robotics |
Online Access: | https://doi.org/10.1049/csy2.12054 |
_version_ | 1811206720038371328 |
---|---|
author | Chi Li Fei Yan Yan Zhuang |
author_facet | Chi Li Fei Yan Yan Zhuang |
author_sort | Chi Li |
collection | DOAJ |
description | Abstract Deep‐learning‐based 3D place recognition has received more attention since the data‐driven fashion is widely used for the 3D point cloud applications. Most of the existing deep‐learning‐based 3D place recognition methods only utilise a single scene for place recognition. However, a single scene may have measurement noise or observable dynamic object differences, which may lead to a reduction in recognition accuracy. To improve the performance of 3D place recognition, a sequence matching based rearrangement method is proposed. Our sequence matching method is based on an assignment algorithm and guides the candidate rearrangement in searching for a similar place. The global descriptor extraction adapts the effective sparse tensor representation and a simple pooling layer to obtain the global descriptor. A new loss function combination is employed to train the network. The proposed approach is evaluated on the popular 3D place recognition benchmarks, which proves the effectiveness of the proposed approach. |
first_indexed | 2024-04-12T03:52:57Z |
format | Article |
id | doaj.art-7e8ca01272fa4e34b0ceff3499571a81 |
institution | Directory Open Access Journal |
issn | 2631-6315 |
language | English |
last_indexed | 2024-04-12T03:52:57Z |
publishDate | 2022-09-01 |
publisher | Wiley |
record_format | Article |
series | IET Cyber-systems and Robotics |
spelling | doaj.art-7e8ca01272fa4e34b0ceff3499571a812022-12-22T03:48:56ZengWileyIET Cyber-systems and Robotics2631-63152022-09-014318919910.1049/csy2.12054Sequence matching enhanced 3D place recognition using candidate rearrangementChi Li0Fei Yan1Yan Zhuang2School of Control Science and Engineering Dalian University of Technology Dalian ChinaSchool of Control Science and Engineering Dalian University of Technology Dalian ChinaSchool of Control Science and Engineering Dalian University of Technology Dalian ChinaAbstract Deep‐learning‐based 3D place recognition has received more attention since the data‐driven fashion is widely used for the 3D point cloud applications. Most of the existing deep‐learning‐based 3D place recognition methods only utilise a single scene for place recognition. However, a single scene may have measurement noise or observable dynamic object differences, which may lead to a reduction in recognition accuracy. To improve the performance of 3D place recognition, a sequence matching based rearrangement method is proposed. Our sequence matching method is based on an assignment algorithm and guides the candidate rearrangement in searching for a similar place. The global descriptor extraction adapts the effective sparse tensor representation and a simple pooling layer to obtain the global descriptor. A new loss function combination is employed to train the network. The proposed approach is evaluated on the popular 3D place recognition benchmarks, which proves the effectiveness of the proposed approach.https://doi.org/10.1049/csy2.12054 |
spellingShingle | Chi Li Fei Yan Yan Zhuang Sequence matching enhanced 3D place recognition using candidate rearrangement IET Cyber-systems and Robotics |
title | Sequence matching enhanced 3D place recognition using candidate rearrangement |
title_full | Sequence matching enhanced 3D place recognition using candidate rearrangement |
title_fullStr | Sequence matching enhanced 3D place recognition using candidate rearrangement |
title_full_unstemmed | Sequence matching enhanced 3D place recognition using candidate rearrangement |
title_short | Sequence matching enhanced 3D place recognition using candidate rearrangement |
title_sort | sequence matching enhanced 3d place recognition using candidate rearrangement |
url | https://doi.org/10.1049/csy2.12054 |
work_keys_str_mv | AT chili sequencematchingenhanced3dplacerecognitionusingcandidaterearrangement AT feiyan sequencematchingenhanced3dplacerecognitionusingcandidaterearrangement AT yanzhuang sequencematchingenhanced3dplacerecognitionusingcandidaterearrangement |