TransFusionOdom: Transformer-based LiDAR-Inertial Fusion Odometry Estimation

Multi-modal fusion of sensors is a commonly used approach to enhance the performance of odometry estimation, which is also a fundamental module for mobile robots. Recently, learning-based approaches garner the attention in this field, due to their robust non-handcrafted designs. However, the questio...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:IEEE sensors journal Ročník 23; číslo 18; s. 1
Hlavní autori: Sun, Leyuan, Ding, Guanqun, Qiu, Yue, Yoshiyasu, Yusuke, Kanehiro, Fumio
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: New York IEEE 15.09.2023
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Institute of Electrical and Electronics Engineers
Predmet:
ISSN:1530-437X, 1558-1748
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Abstract Multi-modal fusion of sensors is a commonly used approach to enhance the performance of odometry estimation, which is also a fundamental module for mobile robots. Recently, learning-based approaches garner the attention in this field, due to their robust non-handcrafted designs. However, the question of How to perform fusion among different modalities in a supervised sensor fusion odometry estimation task? is still one of challenging issues remains. Some simple operations, such as element-wise summation and concatenation, are not capable of assigning adaptive attentional weights to incorporate different modalities efficiently, which make it difficult to achieve competitive odometry results. Besides, the Transformer architecture has shown potential for multi-modal fusion tasks, particularly in the domains of vision with language. In this work, we propose an end-to-end supervised Transformer-based LiDAR-Inertial fusion framework (namely TransFusionOdom) for odometry estimation. The multi-attention fusion module demonstrates different fusion approaches for homogeneous and heterogeneous modalities to address the overfitting problem that can arise from blindly increasing the complexity of the model. Additionally, to interpret the learning process of the Transformer-based multi-modal interactions, a general visualization approach is introduced to illustrate the interactions between modalities. Moreover, exhaustive ablation studies evaluate different multi-modal fusion strategies to verify the performance of proposed fusion strategy. A synthetic multi-modal dataset is made public to validate the generalization ability of the proposed fusion strategy, which also works for other combinations of different modalities. The quantitative and qualitative odometry evaluations on the KITTI dataset verify the proposed TransFusionOdom can achieve superior performance compared with other learning-based related works.
AbstractList Multi-modal fusion of sensors is a commonly used approach to enhance the performance of odometry estimation, which is also a fundamental module for mobile robots. Recently, learning-based approaches garner the attention in this field, due to their robust non-handcrafted designs. However, the question of How to perform fusion among different modalities in a supervised sensor fusion odometry estimation task? is still one of challenging issues remains. Some simple operations, such as element-wise summation and concatenation, are not capable of assigning adaptive attentional weights to incorporate different modalities efficiently, which make it difficult to achieve competitive odometry results. Besides, the Transformer architecture has shown potential for multi-modal fusion tasks, particularly in the domains of vision with language. In this work, we propose an end-to-end supervised Transformer-based LiDAR-Inertial fusion framework (namely TransFusionOdom) for odometry estimation. The multi-attention fusion module demonstrates different fusion approaches for homogeneous and heterogeneous modalities to address the overfitting problem that can arise from blindly increasing the complexity of the model. Additionally, to interpret the learning process of the Transformer-based multi-modal interactions, a general visualization approach is introduced to illustrate the interactions between modalities. Moreover, exhaustive ablation studies evaluate different multi-modal fusion strategies to verify the performance of proposed fusion strategy. A synthetic multi-modal dataset is made public to validate the generalization ability of the proposed fusion strategy, which also works for other combinations of different modalities. The quantitative and qualitative odometry evaluations on the KITTI dataset verify the proposed TransFusionOdom can achieve superior performance compared with other learning-based related works.
Multimodal fusion of sensors is a commonly used approach to enhance the performance of odometry estimation, which is also a fundamental module for mobile robots. Recently, learning-based approaches garner the attention in this field, due to their robust nonhandcrafted designs. However, the question of How to perform fusion among different modalities in a supervised sensor fusion odometry estimation task? is one of the challenging issues still remaining. Some simple operations, such as elementwise summation and concatenation, are not capable of assigning adaptive attentional weights to incorporate different modalities efficiently, which makes it difficult to achieve competitive odometry results. Besides, the Transformer architecture has shown potential for multimodal fusion tasks, particularly in the domains of vision with language. In this work, we propose an end-to-end supervised Transformer-based LiDAR-Inertial fusion framework (namely TransFusionOdom) for odometry estimation. The multiattention fusion module demonstrates different fusion approaches for homogeneous and heterogeneous modalities to address the overfitting problem that can arise from blindly increasing the complexity of the model. Additionally, to interpret the learning process of the Transformer-based multimodal interactions, a general visualization approach is introduced to illustrate the interactions between modalities. Moreover, exhaustive ablation studies evaluate different multimodal fusion strategies to verify the performance of the proposed fusion strategy. A synthetic multimodal dataset is made public to validate the generalization ability of the proposed fusion strategy, which also works for other combinations of different modalities. The quantitative and qualitative odometry evaluations on the KITTI dataset verify that the proposed TransFusionOdom can achieve superior performance compared with other learning-based related works.
Author Qiu, Yue
Sun, Leyuan
Kanehiro, Fumio
Ding, Guanqun
Yoshiyasu, Yusuke
Author_xml – sequence: 1
  givenname: Leyuan
  orcidid: 0000-0001-6123-9339
  surname: Sun
  fullname: Sun, Leyuan
  organization: CNRS-AIST Joint Robotics Laboratory (JRL), IRL, National Institute of Advanced Industrial Science and Technology (AIST), Tsukuba, Ibaraki, Japan
– sequence: 2
  givenname: Guanqun
  surname: Ding
  fullname: Ding, Guanqun
  organization: Digital Architecture Research Center (Di-giARC), National Institute of Advanced Industrial Science and Technology (AIST), Tokyo, Japan
– sequence: 3
  givenname: Yue
  surname: Qiu
  fullname: Qiu, Yue
  organization: Computer Vision Research Team, Artificial Intelligence Research Center (AIRC), National Institute of Advanced Industrial Science and Technology (AIST), Tsukuba, Ibaraki, Japan
– sequence: 4
  givenname: Yusuke
  orcidid: 0000-0002-0433-9832
  surname: Yoshiyasu
  fullname: Yoshiyasu, Yusuke
  organization: Computer Vision Research Team, Artificial Intelligence Research Center (AIRC), National Institute of Advanced Industrial Science and Technology (AIST), Tsukuba, Ibaraki, Japan
– sequence: 5
  givenname: Fumio
  orcidid: 0000-0002-0277-3467
  surname: Kanehiro
  fullname: Kanehiro, Fumio
  organization: CNRS-AIST Joint Robotics Laboratory (JRL), IRL, National Institute of Advanced Industrial Science and Technology (AIST), Tsukuba, Ibaraki, Japan
BackLink https://hal.science/hal-04745599$$DView record in HAL
BookMark eNp9kF9LwzAUxYNMcJt-AMGHgk8-dN40ydL4NubmlOJAJ_gW0jbBjK6dSSfs29uuE8QHn-7l8Dv3zxmgXlmVGqFLDCOMQdw-vc6eRxFEZEQIRBTwCepjxuIQcxr32p5ASAl_P0MD79cAWHDG-yhZOVX6-c7bqlzm1eYuOAimchvtwlR5nQeJvZ-8hI-ldrVVRdDBQUvr2u2Dma_tRtWNdo5OjSq8vjjWIXqbz1bTRZgsHx6nkyTMiKB1qDXkBvOc5qkSIjOGK5amuSDQ6EaMI4UzE2cUC6bzPOYqAgppTDmlOAaKyRDddHM_VCG3rtnu9rJSVi4miWw1aFjGhPhq2euO3brqc6d9LdfVzpXNeTKKxwwEMAENxTsqc5X3ThuZ2frwU-2ULSQG2cYs25hlG7M8xtw48R_nz0H_ea46j9Va_-IjTBkek29SlInS
CODEN ISJEAZ
CitedBy_id crossref_primary_10_1016_j_robot_2024_104700
crossref_primary_10_1109_JSEN_2025_3586680
crossref_primary_10_1016_j_rineng_2024_103565
crossref_primary_10_1016_j_aei_2025_103135
crossref_primary_10_3390_s25061839
crossref_primary_10_1016_j_isprsjprs_2024_03_008
crossref_primary_10_1109_JSEN_2024_3442951
crossref_primary_10_1109_JSEN_2025_3562916
crossref_primary_10_3389_fcomp_2024_1486581
crossref_primary_10_1016_j_patcog_2025_111383
crossref_primary_10_1109_JSEN_2025_3530007
crossref_primary_10_1109_JSEN_2024_3386709
crossref_primary_10_1007_s10462_025_11187_w
crossref_primary_10_3390_drones7120699
Cites_doi 10.1109/IROS47612.2022.9981107
10.1109/ICRA48506.2021.9561149
10.1109/TRO.2018.2853729
10.1109/LRA.2022.3141661
10.1109/TCE.2003.1233761
10.1109/CVPR.2019.00867
10.1109/CVPR46437.2021.00700
10.1109/ICIP40778.2020.9190664
10.1109/ISMAR.2014.6948420
10.1109/LRA.2020.3003256
10.1021/ac60214a047
10.3390/rs14205229
10.1109/TRO.2017.2705103
10.1109/ICCV.2015.336
10.1109/JSEN.2019.2910826
10.1109/CVPR52729.2023.02174
10.1007/978-3-031-02444-3_14
10.1109/CVPR52688.2022.01178
10.1016/j.neunet.2022.09.001
10.1109/ICCV48922.2021.00273
10.1109/CVPR52688.2022.01187
10.1109/ROBIO55434.2022.10011808
10.1109/CVPR.2019.01079
10.1109/ICASSP39728.2021.9413912
10.1109/IROS45743.2020.9341176
10.1109/JSEN.2020.3028561
10.1109/TAES.2022.3193085
10.1109/IROS47612.2022.9981835
10.1109/ICRA.2019.8793511
10.1109/ICRA46639.2022.9811842
10.1007/s12555-020-0443-2
10.1007/978-3-030-66498-5_24
10.1109/ICRA46639.2022.9812027
10.1109/ICRA.2017.7989236
10.1109/IROS.2018.8594299
10.1109/TNNLS.2022.3176677
10.1109/JSEN.2019.2947446
10.1609/aaai.v31i1.11215
10.1109/ICCV48922.2021.00299
10.1109/CVPR.2018.00781
10.1177/0020294019858217
10.1109/LRA.2021.3095515
10.1109/CVPR.2016.90
10.1007/978-3-030-58568-6_42
10.1016/j.measurement.2022.111030
10.1109/TIV.2023.3273288
10.1109/TRO.2022.3141876
10.1109/CVPR52688.2022.00475
10.1109/IROS40897.2019.8967880
10.1109/ICRA48506.2021.9560947
10.1109/TPAMI.2023.3275156/mm1
10.15607/RSS.2009.V.021
10.1109/JIOT.2022.3151629
10.1371/journal.pone.0261053
10.1109/IROS40897.2019.8967762
10.5194/isprs-annals-V-1-2021-47-2021
10.1109/ICRA40945.2020.9197366
10.1109/LRA.2021.3064227
10.1109/CVPR.2012.6248074
10.1109/JSEN.2021.3128683
10.1016/j.patcog.2020.107618
10.1109/34.121791
10.1109/JSEN.2022.3208200
10.1109/ICCV48922.2021.00041
10.1109/CVPR.2019.00589
10.15607/RSS.2014.X.007
ContentType Journal Article
Copyright Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2023
Distributed under a Creative Commons Attribution 4.0 International License
Copyright_xml – notice: Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2023
– notice: Distributed under a Creative Commons Attribution 4.0 International License
DBID 97E
RIA
RIE
AAYXX
CITATION
7SP
7U5
8FD
L7M
1XC
DOI 10.1109/JSEN.2023.3302401
DatabaseName IEEE Xplore (IEEE)
IEEE All-Society Periodicals Package (ASPP) 1998–Present
IEEE Electronic Library (IEL)
CrossRef
Electronics & Communications Abstracts
Solid State and Superconductivity Abstracts
Technology Research Database
Advanced Technologies Database with Aerospace
Hyper Article en Ligne (HAL)
DatabaseTitle CrossRef
Solid State and Superconductivity Abstracts
Technology Research Database
Advanced Technologies Database with Aerospace
Electronics & Communications Abstracts
DatabaseTitleList
Solid State and Superconductivity Abstracts
Database_xml – sequence: 1
  dbid: RIE
  name: IEEE Electronic Library (IEL)
  url: https://ieeexplore.ieee.org/
  sourceTypes: Publisher
DeliveryMethod fulltext_linktorsrc
Discipline Geography
Engineering
Computer Science
EISSN 1558-1748
EndPage 1
ExternalDocumentID oai:HAL:hal-04745599v1
10_1109_JSEN_2023_3302401
10214516
Genre orig-research
GrantInformation_xml – fundername: JST-SPRING
  grantid: JPMJSP2124
– fundername: JSPS KAKENHI
  grantid: 23H03426
GroupedDBID -~X
0R~
29I
4.4
5GY
6IK
97E
AAJGR
AARMG
AASAJ
AAWTH
ABAZT
ABQJQ
ABVLG
ACGFO
ACGFS
ACIWK
AENEX
AGQYO
AHBIQ
AJQPL
AKJIK
AKQYR
ALMA_UNASSIGNED_HOLDINGS
ATWAV
BEFXN
BFFAM
BGNUA
BKEBE
BPEOZ
CS3
EBS
F5P
HZ~
IFIPE
IPLJI
JAVBF
LAI
M43
O9-
OCL
P2P
RIA
RIE
RNS
TWZ
AAYXX
CITATION
7SP
7U5
8FD
L7M
1XC
ID FETCH-LOGICAL-c394t-ee0df17d4dba99cff7a5bbd9300dff962a1cf8c4195edd87a2040b84744180413
IEDL.DBID RIE
ISICitedReferencesCount 16
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001090399700160&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 1530-437X
IngestDate Tue Oct 14 20:21:22 EDT 2025
Mon Jun 30 10:09:52 EDT 2025
Tue Nov 18 20:45:57 EST 2025
Sat Nov 29 06:39:39 EST 2025
Mon Aug 04 05:48:55 EDT 2025
IsPeerReviewed true
IsScholarly true
Issue 18
Language English
License https://ieeexplore.ieee.org/Xplorehelp/downloads/license-information/IEEE.html
https://doi.org/10.15223/policy-029
https://doi.org/10.15223/policy-037
Distributed under a Creative Commons Attribution 4.0 International License: http://creativecommons.org/licenses/by/4.0
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-c394t-ee0df17d4dba99cff7a5bbd9300dff962a1cf8c4195edd87a2040b84744180413
Notes ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ORCID 0000-0002-0433-9832
0000-0002-0277-3467
0000-0001-6123-9339
PQID 2865090590
PQPubID 75733
PageCount 1
ParticipantIDs hal_primary_oai_HAL_hal_04745599v1
proquest_journals_2865090590
crossref_primary_10_1109_JSEN_2023_3302401
ieee_primary_10214516
crossref_citationtrail_10_1109_JSEN_2023_3302401
PublicationCentury 2000
PublicationDate 2023-09-15
PublicationDateYYYYMMDD 2023-09-15
PublicationDate_xml – month: 09
  year: 2023
  text: 2023-09-15
  day: 15
PublicationDecade 2020
PublicationPlace New York
PublicationPlace_xml – name: New York
PublicationTitle IEEE sensors journal
PublicationTitleAbbrev JSEN
PublicationYear 2023
Publisher IEEE
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Institute of Electrical and Electronics Engineers
Publisher_xml – name: IEEE
– name: The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
– name: Institute of Electrical and Electronics Engineers
References ref13
chen (ref29) 2020
ref12
ref56
ref59
ref14
ref58
ref53
ref52
ref55
ref10
ref54
ref16
kim (ref42) 2021
ref19
ref18
dosovitskiy (ref57) 2020
wang (ref35) 2022
wang (ref11) 2020; 33
ref51
ref50
ref45
ref48
zhang (ref77) 2021
ref47
ref41
ref44
ref43
huang (ref80) 2023
vaswani (ref23) 2017; 30
ref49
ref8
ref7
ref9
ref4
ref3
ref6
xu (ref22) 2023
ref5
ref40
ref79
cho (ref15) 2019
ref34
ref78
ref37
cao (ref17) 2020
ref36
kendall (ref69) 2017; 30
ref31
ref75
ref30
ref74
ref33
ref32
liu (ref76) 2021; 34
ref2
ref1
ref39
ref38
chen (ref64) 2022
ref71
ref70
ref73
ref72
hassani (ref46) 2021
ref24
ref68
ref67
ref26
ref25
ref20
ref63
ref66
ref21
ref65
ref28
ref27
ref60
ref62
ref61
References_xml – start-page: 5583
  year: 2021
  ident: ref42
  article-title: ViLT: Vision-and-language transformer without convolution or region supervision
  publication-title: Proc Int Conf Mach Learn (ICML)
– year: 2019
  ident: ref15
  article-title: DeepLO: Geometry-aware deep LiDAR odometry
  publication-title: arXiv 1902 10562 Icsl
– ident: ref39
  doi: 10.1109/IROS47612.2022.9981107
– ident: ref41
  doi: 10.1109/ICRA48506.2021.9561149
– ident: ref18
  doi: 10.1109/TRO.2018.2853729
– ident: ref79
  doi: 10.1109/LRA.2022.3141661
– ident: ref65
  doi: 10.1109/TCE.2003.1233761
– ident: ref51
  doi: 10.1109/CVPR.2019.00867
– ident: ref60
  doi: 10.1109/CVPR46437.2021.00700
– year: 2020
  ident: ref17
  article-title: Fast monocular visual odometry for augmented reality on smartphones
  publication-title: IEEE Consum Electron Mag
– ident: ref71
  doi: 10.1109/ICIP40778.2020.9190664
– ident: ref16
  doi: 10.1109/ISMAR.2014.6948420
– ident: ref55
  doi: 10.1109/LRA.2020.3003256
– ident: ref56
  doi: 10.1021/ac60214a047
– ident: ref74
  doi: 10.3390/rs14205229
– ident: ref27
  doi: 10.1109/TRO.2017.2705103
– ident: ref66
  doi: 10.1109/ICCV.2015.336
– ident: ref8
  doi: 10.1109/JSEN.2019.2910826
– ident: ref24
  doi: 10.1109/CVPR52729.2023.02174
– ident: ref49
  doi: 10.1007/978-3-031-02444-3_14
– ident: ref63
  doi: 10.1109/CVPR52688.2022.01178
– volume: 33
  start-page: 4835
  year: 2020
  ident: ref11
  article-title: Deep multimodal fusion by channel exchanging
  publication-title: Proc Adv Neural Inf Process Syst
– ident: ref10
  doi: 10.1016/j.neunet.2022.09.001
– ident: ref68
  doi: 10.1109/ICCV48922.2021.00273
– ident: ref44
  doi: 10.1109/CVPR52688.2022.01187
– ident: ref14
  doi: 10.1109/ROBIO55434.2022.10011808
– ident: ref3
  doi: 10.1109/CVPR.2019.01079
– ident: ref6
  doi: 10.1109/ICASSP39728.2021.9413912
– ident: ref19
  doi: 10.1109/IROS45743.2020.9341176
– ident: ref54
  doi: 10.1109/JSEN.2020.3028561
– ident: ref78
  doi: 10.1109/TAES.2022.3193085
– ident: ref2
  doi: 10.1109/IROS47612.2022.9981835
– ident: ref38
  doi: 10.1109/ICRA.2019.8793511
– ident: ref45
  doi: 10.1109/ICRA46639.2022.9811842
– ident: ref4
  doi: 10.1007/s12555-020-0443-2
– year: 2021
  ident: ref46
  article-title: Escaping the big data paradigm with compact transformers
  publication-title: arXiv 2104 05704
– ident: ref52
  doi: 10.1007/978-3-030-66498-5_24
– volume: 30
  start-page: 1
  year: 2017
  ident: ref23
  article-title: Attention is all you need
  publication-title: Proc Adv Neural Inf Process Syst
– ident: ref43
  doi: 10.1109/ICRA46639.2022.9812027
– ident: ref67
  doi: 10.1109/ICRA.2017.7989236
– ident: ref32
  doi: 10.1109/IROS.2018.8594299
– ident: ref21
  doi: 10.1109/TNNLS.2022.3176677
– ident: ref53
  doi: 10.1109/JSEN.2019.2947446
– ident: ref7
  doi: 10.1609/aaai.v31i1.11215
– ident: ref61
  doi: 10.1109/ICCV48922.2021.00299
– ident: ref72
  doi: 10.1109/CVPR.2018.00781
– year: 2020
  ident: ref57
  article-title: An image is worth 16 × 16 words: Transformers for image recognition at scale
  publication-title: arXiv 2010 11929
– ident: ref50
  doi: 10.1177/0020294019858217
– ident: ref40
  doi: 10.1109/LRA.2021.3095515
– year: 2020
  ident: ref29
  article-title: A survey on deep learning for localization and mapping: Towards the age of spatial machine intelligence
  publication-title: arXiv 2006 12567
– volume: 30
  start-page: 1
  year: 2017
  ident: ref69
  article-title: What uncertainties do we need in Bayesian deep learning for computer vision?
  publication-title: Proc Adv Neural Inf Process Syst
– volume: 34
  start-page: 23818
  year: 2021
  ident: ref76
  article-title: Efficient training of visual transformers with small datasets
  publication-title: Proc Adv Neural Inf Process Syst
– ident: ref47
  doi: 10.1109/CVPR.2016.90
– ident: ref70
  doi: 10.1007/978-3-030-58568-6_42
– ident: ref9
  doi: 10.1016/j.measurement.2022.111030
– ident: ref34
  doi: 10.1109/TIV.2023.3273288
– ident: ref37
  doi: 10.1109/TRO.2022.3141876
– ident: ref58
  doi: 10.1109/CVPR52688.2022.00475
– ident: ref33
  doi: 10.1109/IROS40897.2019.8967880
– ident: ref73
  doi: 10.1109/ICRA48506.2021.9560947
– year: 2023
  ident: ref22
  article-title: Multimodal learning with transformers: A survey
  publication-title: IEEE Trans Pattern Anal Mach Intell
  doi: 10.1109/TPAMI.2023.3275156/mm1
– ident: ref31
  doi: 10.15607/RSS.2009.V.021
– year: 2022
  ident: ref35
  article-title: D-LIOM: Tightly-coupled direct LiDAR-inertial odometry and mapping
  publication-title: IEEE Trans Multimedia
– ident: ref26
  doi: 10.1109/JIOT.2022.3151629
– ident: ref20
  doi: 10.1371/journal.pone.0261053
– year: 2022
  ident: ref64
  article-title: CF-ViT: A general coarse-to-fine method for vision transformer
  publication-title: arXiv 2203 03821
– ident: ref48
  doi: 10.1109/IROS40897.2019.8967762
– ident: ref5
  doi: 10.5194/isprs-annals-V-1-2021-47-2021
– ident: ref59
  doi: 10.1109/ICRA40945.2020.9197366
– ident: ref36
  doi: 10.1109/LRA.2021.3064227
– ident: ref13
  doi: 10.1109/CVPR.2012.6248074
– ident: ref25
  doi: 10.1109/JSEN.2021.3128683
– ident: ref12
  doi: 10.1016/j.patcog.2020.107618
– start-page: 22690
  year: 2023
  ident: ref80
  article-title: Vision transformer with super token sampling
  publication-title: Proc Conf Comput Vis Pattern Recognit (CVPR)
– ident: ref30
  doi: 10.1109/34.121791
– year: 2021
  ident: ref77
  article-title: MixUp training leads to reduced overfitting and improved calibration for the transformer architecture
  publication-title: arXiv 2102 11402
– ident: ref1
  doi: 10.1109/JSEN.2022.3208200
– ident: ref62
  doi: 10.1109/ICCV48922.2021.00041
– ident: ref75
  doi: 10.1109/CVPR.2019.00589
– ident: ref28
  doi: 10.15607/RSS.2014.X.007
SSID ssj0019757
Score 2.4738455
Snippet Multi-modal fusion of sensors is a commonly used approach to enhance the performance of odometry estimation, which is also a fundamental module for mobile...
Multimodal fusion of sensors is a commonly used approach to enhance the performance of odometry estimation, which is also a fundamental module for mobile...
SourceID hal
proquest
crossref
ieee
SourceType Open Access Repository
Aggregation Database
Enrichment Source
Index Database
Publisher
StartPage 1
SubjectTerms Ablation
Attention mechanisms
Computer Science
Datasets
Estimation
Inertial fusion (reactor)
Laser radar
Learning
Lidar
LiDAR-inertial odometry
Modules
multi-modal learning
Multisensor fusion
Odometry
Robotics
sensor data fusion
Sensor fusion
Sensors
Strategy
Task analysis
transformer Vertex & Normal
Transformers
Title TransFusionOdom: Transformer-based LiDAR-Inertial Fusion Odometry Estimation
URI https://ieeexplore.ieee.org/document/10214516
https://www.proquest.com/docview/2865090590
https://hal.science/hal-04745599
Volume 23
WOSCitedRecordID wos001090399700160&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
journalDatabaseRights – providerCode: PRVIEE
  databaseName: IEEE Electronic Library (IEL)
  customDbUrl:
  eissn: 1558-1748
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0019757
  issn: 1530-437X
  databaseCode: RIE
  dateStart: 20010101
  isFulltext: true
  titleUrlDefault: https://ieeexplore.ieee.org/
  providerName: IEEE
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV3fa9swED6WUtj60HZdR9Mfw4w9DZRKsWVZfQttQjdCNrYO8mYs6UQLa1LSJJD_vjrZCR1jhb0Z-SSEP0ufftzdB_ApkwaFtAXz1nmWVdKxqlKOCeUDP0jj08xGsQk1GhXjsf7eBKvHWBhEjM5n2KHHeJfvpnZBR2XnUYZairwFLaXyOlhrc2WgVUzrGUYwZ1mqxs0VpuD6_OvP_qhDOuGdsHsPFCb-IKHWLblARm2VvybkyDKDvf_s3z7sNsvJpFfj_xZe4eQAdp4lGTyA143O-e3qHQwjNQ0WdET2zU3vL5Kb9cIVZ4wYzSXDu6veD_ZlQg7XoenaOCFrnM9WST_MCXW44yH8GvRvLq9Zo6fAbKqzOUPkzgvlMmcqra33qpLGOJ3yUO513q2E9YXNhJboXKGqbhjhJtBXWDJRmqL0PWxNphM8ggSzqihIOd0Eeg8vTc67zhddlKilybENfP2BS9skGyfNi99l3HRwXRImJWFSNpi04fOmykOdaeMl448BtY0d5ci-7g1LKuOhv5RGbRmMDgmjZ63V8LThdI1y2YzYx5IidLmmUNzjf1Q7gTfUBXIWEfIUtuazBZ7Btl3O7x5nH-LP-AQaUtq8
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV3fixMxEB68Uzh98Md5YvXURXwS0kt2k-7Gt6ItPV2raIW-hU0y4Q60lV57cP-9mey2nIiCb0t2EsJOki-bzHwfwCupLArlKhacD0w2yrOmKT0TZYj4oGwopEtiE-V0Ws3n-nOXrJ5yYRAxBZ9hnx7TXb5fug0dlZ0kGWolBntwU0mZ8zZda3dpoMtE7BnnMGeyKOfdJabg-uT919G0T0rh_fj_HkFM_AZDe2cUBJnUVf5YkhPOjO_9Zw_vw91uQ5kN2xHwAG7g4hDuXKMZPISDTun87Ooh1Amcxhs6JPvklz_eZLPt1hVXjDDNZ_X5u-EXdrqgkOvYdGuckTWuV1fZKK4KbcLjEXwbj2ZvJ6xTVGCu0HLNELkPovTS20ZrF0LZKGu9LngsD3qQN8KFykmhFXpflU0e57iNABY3TURUVDyC_cVygY8hQ9lUFWmn2wjw8aUd8NyHKkeFWtkB9oBvP7BxHd04qV58N-m3g2tDPjHkE9P5pAevd1V-tlwb_zJ-Gb22syOW7MmwNlTGY3-JSO0yGh2Rj6611rqnB8dbL5tuzl4YytHlmpJxn_yl2gs4mMw-1qY-nX54CrepOxQ6ItQx7K9XG3wGt9zl-vxi9TwNzF-knN4D
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=TransFusionOdom%3A+Transformer-Based+LiDAR-Inertial+Fusion+Odometry+Estimation&rft.jtitle=IEEE+sensors+journal&rft.au=Sun%2C+Leyuan&rft.au=Ding%2C+Guanqun&rft.au=Qiu%2C+Yue&rft.au=Yoshiyasu%2C+Yusuke&rft.date=2023-09-15&rft.pub=Institute+of+Electrical+and+Electronics+Engineers&rft.issn=1530-437X&rft.volume=23&rft.issue=18&rft.spage=22064&rft.epage=22079&rft_id=info:doi/10.1109%2FJSEN.2023.3302401&rft.externalDBID=HAS_PDF_LINK&rft.externalDocID=oai%3AHAL%3Ahal-04745599v1
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1530-437X&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1530-437X&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1530-437X&client=summon