Representation of Multirelations of Geographic Scenes Based on Hyperbolic Space

The recognition of complex geographic scenes requires to simultaneously describe different kinds of relations such as hyponymy and spatial relations among multiple-scale categories. The former is hierarchical structure, and the latter is called eigen structure. Current researches focus on representi...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on geoscience and remote sensing Vol. 62; pp. 1 - 13
Main Authors: Cui, Wei, Xu, Xing, Chen, Jiale, Feng, Zhanyun, Zhao, Huilin, Hao, Yuanjie, Tian, Yueling, Wang, Jin, Xia, Cong
Format: Journal Article
Language:English
Published: New York IEEE 2024
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Subjects:
Artificial neural networks
Computational modeling
Embedding
Euclidean geometry
Euclidean space
Feature extraction
Geographic scenes
Geometric transformation
Graph representations
Graphical representations
Hyperbolic coordinates
hyperbolic space
Knowledge graphs
Knowledge representation
Machine learning
multirelations
Neural networks
Object recognition
Quaternions
Remote sensing
Semantic segmentation
Semantics
Vectors
Visualization
ISSN:0196-2892, 1558-0644
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Abstract The recognition of complex geographic scenes requires to simultaneously describe different kinds of relations such as hyponymy and spatial relations among multiple-scale categories. The former is hierarchical structure, and the latter is called eigen structure. Current researches focus on representing these relations in Euclidean space. However, the key characteristic of the Euclidean spaces is that it expands polynomially with respect to the radius, which makes it difficult to describe both the hierarchical and eigen structure within finite computable dimensions simultaneously. To address the above issues, a hyperbolic embedding model for representing complex geographic scenes (HEMR) is first proposed. First, the visual features of remote sensing objects are extracted through a deep neural network in Euclidean space and mapped onto the n-dimensional Poincare ball in hyperbolic space by an exponential projection. Second, a two-stage contrastive learning mechanism based on Möbius transformation is designed, which uses the property of exponential expansion to obtain the ability to represent different kinds of relations. Finally, a new quaternion loss is designed to describe the relations between feature categories in remote sensing scenarios by hyperbolic distance. The experiments indicate that our model can leverage the characteristic of spatial exponential growth with the radius in hyperbolic space, simultaneously describing the hierarchical and eigen structures of complex remote sensing scenes in finite dimensions. This provides support for downstream tasks such as semantic segmentation of remote sensing, knowledge graph representation, and link prediction.
AbstractList The recognition of complex geographic scenes requires to simultaneously describe different kinds of relations such as hyponymy and spatial relations among multiple-scale categories. The former is hierarchical structure, and the latter is called eigen structure. Current researches focus on representing these relations in Euclidean space. However, the key characteristic of the Euclidean spaces is that it expands polynomially with respect to the radius, which makes it difficult to describe both the hierarchical and eigen structure within finite computable dimensions simultaneously. To address the above issues, a hyperbolic embedding model for representing complex geographic scenes (HEMR) is first proposed. First, the visual features of remote sensing objects are extracted through a deep neural network in Euclidean space and mapped onto the n-dimensional Poincare ball in hyperbolic space by an exponential projection. Second, a two-stage contrastive learning mechanism based on Möbius transformation is designed, which uses the property of exponential expansion to obtain the ability to represent different kinds of relations. Finally, a new quaternion loss is designed to describe the relations between feature categories in remote sensing scenarios by hyperbolic distance. The experiments indicate that our model can leverage the characteristic of spatial exponential growth with the radius in hyperbolic space, simultaneously describing the hierarchical and eigen structures of complex remote sensing scenes in finite dimensions. This provides support for downstream tasks such as semantic segmentation of remote sensing, knowledge graph representation, and link prediction.
Author Cui, Wei
Zhao, Huilin
Tian, Yueling
Hao, Yuanjie
Wang, Jin
Xu, Xing
Chen, Jiale
Feng, Zhanyun
Xia, Cong
Author_xml – sequence: 1
  givenname: Wei
  surname: Cui
  fullname: Cui, Wei
  email: cuiwei@whut.edu.cn
  organization: School of Resource and Environmental Engineering, Wuhan University of Technology, Wuhan, China
– sequence: 2
  givenname: Xing
  orcidid: 0000-0002-2310-4385
  surname: Xu
  fullname: Xu, Xing
  email: xuxing@whut.edu.cn
  organization: School of Resource and Environmental Engineering, Wuhan University of Technology, Wuhan, China
– sequence: 3
  givenname: Jiale
  surname: Chen
  fullname: Chen, Jiale
  email: chenjl07@whut.edu.cn
  organization: School of Resource and Environmental Engineering, Wuhan University of Technology, Wuhan, China
– sequence: 4
  givenname: Zhanyun
  surname: Feng
  fullname: Feng, Zhanyun
  email: 276762@whut.edu.cn
  organization: School of Resource and Environmental Engineering, Wuhan University of Technology, Wuhan, China
– sequence: 5
  givenname: Huilin
  orcidid: 0000-0002-2843-1123
  surname: Zhao
  fullname: Zhao, Huilin
  email: zhaohl2016@whut.edu.cn
  organization: School of Resource and Environmental Engineering, Wuhan University of Technology, Wuhan, China
– sequence: 6
  givenname: Yuanjie
  surname: Hao
  fullname: Hao, Yuanjie
  email: haoyuanjie@whut.edu.cn
  organization: School of Resource and Environmental Engineering, Wuhan University of Technology, Wuhan, China
– sequence: 7
  givenname: Yueling
  surname: Tian
  fullname: Tian, Yueling
  email: tianyueling@whut.edu.cn
  organization: School of Resource and Environmental Engineering, Wuhan University of Technology, Wuhan, China
– sequence: 8
  givenname: Jin
  surname: Wang
  fullname: Wang, Jin
  email: 0121608900228@whut.edu.cn
  organization: School of Resource and Environmental Engineering, Wuhan University of Technology, Wuhan, China
– sequence: 9
  givenname: Cong
  surname: Xia
  fullname: Xia, Cong
  email: 265107@whut.edu.cn
  organization: School of Resource and Environmental Engineering, Wuhan University of Technology, Wuhan, China
BookMark eNpNkE1LAzEQhoNUsK3-AMHDguetyeb7qEVboVJo6zlksxPdUjdrsj3037trPXiaYeZ5Z-CZoFETGkDoluAZIVg_7Bab7azABZtRhqVi9AKNCecqx4KxERpjokVeKF1coUlKe4wJ40SO0XoDbYQETWe7OjRZ8Nnb8dDVEQ6_gzRMFhA-om0_a5dtHTSQsieboMp6fnlqIZbhMKxa6-AaXXp7SHDzV6fo_eV5N1_mq_Xidf64yl3BRJcr62mpASqvnVLUl7LvS6LKqvBcKK0cFBUjQnvPieUcWyuJJN55LAUtKzpF9-e7bQzfR0id2YdjbPqXhmIhNGGS8Z4iZ8rFkFIEb9pYf9l4MgSbwZsZvJnBm_nz1mfuzpkaAP7xnFEsGf0Br-lsYA
CODEN IGRSD2
Cites_doi 10.3390/rs13071312
10.1016/j.isprsjprs.2022.01.009
10.1609/aaai.v28i1.8870
10.48550/ARXIV.1609.02907
10.1145/3583780.3614979
10.1145/2623330.2623732
10.1016/j.rse.2022.112916
10.3390/rs15040880
10.1109/TAC.2013.2254619
10.3390/rs11010083
10.3390/ijgi10010032
10.1109/CSPC.2017.8305844
10.1016/j.neucom.2022.04.026
10.1016/j.ins.2022.07.060
10.1609/aaai.v29i1.9491
10.1109/MSP.2017.2693418
10.1016/j.inffus.2023.101909
10.1145/2939672.2939754
10.1109/tbdata.2021.3081431
10.1109/CVPR.2015.7298682
10.3390/rs9030243
10.1109/TGRS.2020.3016020
10.3390/rs14194872
10.18653/v1/2020.acl-main.617
10.1109/TGRS.2021.3073269
10.3390/rs14051161
10.1016/j.isprsjprs.2021.03.016
10.3390/rs13214201
ContentType Journal Article
Copyright Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2024
Copyright_xml – notice: Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2024
DBID 97E
ESBDL
RIA
RIE
AAYXX
CITATION
7UA
8FD
C1K
F1W
FR3
H8D
H96
KR7
L.G
L7M
DOI 10.1109/TGRS.2024.3407843
DatabaseName IEEE All-Society Periodicals Package (ASPP) 2005–Present
IEEE Xplore Open Access Journals
IEEE All-Society Periodicals Package (ASPP) 1998–Present
IEEE/IET Electronic Library (IEL) (UW System Shared)
CrossRef
Water Resources Abstracts
Technology Research Database
Environmental Sciences and Pollution Management
ASFA: Aquatic Sciences and Fisheries Abstracts
Engineering Research Database
Aerospace Database
Aquatic Science & Fisheries Abstracts (ASFA) 2: Ocean Technology, Policy & Non-Living Resources
Civil Engineering Abstracts
Aquatic Science & Fisheries Abstracts (ASFA) Professional
Advanced Technologies Database with Aerospace
DatabaseTitle CrossRef
Aerospace Database
Civil Engineering Abstracts
Aquatic Science & Fisheries Abstracts (ASFA) Professional
Aquatic Science & Fisheries Abstracts (ASFA) 2: Ocean Technology, Policy & Non-Living Resources
Technology Research Database
ASFA: Aquatic Sciences and Fisheries Abstracts
Engineering Research Database
Advanced Technologies Database with Aerospace
Water Resources Abstracts
Environmental Sciences and Pollution Management
DatabaseTitleList
Aerospace Database
Database_xml – sequence: 1
  dbid: RIE
  name: IEEE/IET Electronic Library (IEL) (UW System Shared)
  url: https://ieeexplore.ieee.org/
  sourceTypes: Publisher
DeliveryMethod fulltext_linktorsrc
Discipline Engineering
Physics
EISSN 1558-0644
EndPage 13
ExternalDocumentID 10_1109_TGRS_2024_3407843
10543074
Genre orig-research
GrantInformation_xml – fundername: National Key Research and Development Program of China
  grantid: 2018YFC0810605
  funderid: 10.13039/501100012166
GroupedDBID -~X
0R~
29I
4.4
5GY
5VS
6IK
97E
AAJGR
AARMG
AASAJ
AAWTH
ABAZT
ABQJQ
ABVLG
ACGFO
ACGFS
ACIWK
ACNCT
AENEX
AETIX
AFRAH
AGQYO
AGSQL
AHBIQ
AI.
AIBXA
AKJIK
AKQYR
ALLEH
ALMA_UNASSIGNED_HOLDINGS
ASUFR
ATWAV
BEFXN
BFFAM
BGNUA
BKEBE
BPEOZ
CS3
DU5
EBS
EJD
ESBDL
F5P
HZ~
H~9
IBMZZ
ICLAB
IFIPE
IFJZH
IPLJI
JAVBF
LAI
M43
O9-
OCL
P2P
RIA
RIE
RNS
RXW
TAE
TN5
VH1
Y6R
AAYXX
CITATION
7UA
8FD
C1K
F1W
FR3
H8D
H96
KR7
L.G
L7M
ID FETCH-LOGICAL-c246t-8af3b9eedf9c883fb7eedb18bd2f56898ce2d4169ff51a550aa7171fcf0763bd3
IEDL.DBID RIE
ISICitedReferencesCount 3
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001246585400047&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 0196-2892
IngestDate Mon Jun 30 08:28:28 EDT 2025
Sat Nov 29 03:32:36 EST 2025
Wed Aug 27 02:15:44 EDT 2025
IsDoiOpenAccess true
IsOpenAccess true
IsPeerReviewed true
IsScholarly true
Language English
License https://creativecommons.org/licenses/by-nc-nd/4.0
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-c246t-8af3b9eedf9c883fb7eedb18bd2f56898ce2d4169ff51a550aa7171fcf0763bd3
Notes ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ORCID 0000-0002-2310-4385
0000-0002-2843-1123
OpenAccessLink https://ieeexplore.ieee.org/document/10543074
PQID 3066914745
PQPubID 85465
PageCount 13
ParticipantIDs ieee_primary_10543074
crossref_primary_10_1109_TGRS_2024_3407843
proquest_journals_3066914745
PublicationCentury 2000
PublicationDate 20240000
2024-00-00
20240101
PublicationDateYYYYMMDD 2024-01-01
PublicationDate_xml – year: 2024
  text: 20240000
PublicationDecade 2020
PublicationPlace New York
PublicationPlace_xml – name: New York
PublicationTitle IEEE transactions on geoscience and remote sensing
PublicationTitleAbbrev TGRS
PublicationYear 2024
Publisher IEEE
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Publisher_xml – name: IEEE
– name: The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
References ref13
Hamilton (ref7); 30
ref12
ref15
ref37
ref14
ref31
ref30
ref11
ref33
ref10
ref32
ref2
Dosovitskiy (ref36) 2020
ref1
ref39
ref38
Hian Lee (ref24) 2023
ref18
Deng (ref23) 2023
Nickel (ref17)
Yang (ref20) 2022
Velickovic (ref6) 2017; 1050
Gu (ref35)
ref25
ref22
ref21
ref28
ref27
Balazevic (ref16); 32
Nickel (ref19); 80
ref29
ref8
ref9
Bai (ref26) 2021
ref4
ref3
Ganea (ref34)
ref5
ref40
McInnes (ref41) 2018
References_xml – ident: ref11
  doi: 10.3390/rs13071312
– ident: ref13
  doi: 10.1016/j.isprsjprs.2022.01.009
– ident: ref1
  doi: 10.1609/aaai.v28i1.8870
– ident: ref5
  doi: 10.48550/ARXIV.1609.02907
– start-page: 6341
  volume-title: Proc. 31st Int. Conf. Neural Inf. Process. Syst.
  ident: ref17
  article-title: Poincaré embeddings for learning hierarchical representations
– year: 2020
  ident: ref36
  article-title: An image is worth 16×16 words: Transformers for image recognition at scale
  publication-title: arXiv:2010.11929
– volume: 30
  start-page: 1025
  volume-title: Proc. Adv. Neural Inf. Process. Syst.
  ident: ref7
  article-title: Inductive representation learning on large graphs
– volume: 80
  start-page: 3776
  volume-title: Proc. 35th Int. Conf. Mach. Learn.
  ident: ref19
  article-title: Learning continuous hierarchies in the Lorentz model of hyperbolic geometry
– ident: ref25
  doi: 10.1145/3583780.3614979
– ident: ref3
  doi: 10.1145/2623330.2623732
– ident: ref9
  doi: 10.1016/j.rse.2022.112916
– ident: ref31
  doi: 10.3390/rs15040880
– start-page: 1
  volume-title: Proc. Int. Conf. Learn. Represent.
  ident: ref35
  article-title: Learning mixed-curvature representations in product spaces
– year: 2023
  ident: ref24
  article-title: Node-specific space selection via localized geometric hyperbolicity in graph neural networks
  publication-title: arXiv:2303.01724
– ident: ref39
  doi: 10.1109/TAC.2013.2254619
– start-page: 1646
  volume-title: Proc. Int. Conf. Mach. Learn.
  ident: ref34
  article-title: Hyperbolic entailment cones for learning hierarchical embeddings
– ident: ref14
  doi: 10.3390/rs11010083
– volume: 32
  start-page: 1
  volume-title: Proc. Adv. Neural Inf. Process. Syst.
  ident: ref16
  article-title: Multi-relational Poincaré graph embeddings
– year: 2021
  ident: ref26
  article-title: Modeling heterogeneous hierarchies with relation-specific hyperbolic cones
  publication-title: arXiv:2110.14923
– year: 2022
  ident: ref20
  article-title: Hyperbolic curvature graph neural network
  publication-title: arXiv:2212.01793
– year: 2023
  ident: ref23
  article-title: FMGNN: Fused manifold graph neural network
  publication-title: arXiv:2304.01081
– ident: ref32
  doi: 10.3390/ijgi10010032
– ident: ref40
  doi: 10.1109/CSPC.2017.8305844
– ident: ref21
  doi: 10.1016/j.neucom.2022.04.026
– ident: ref22
  doi: 10.1016/j.ins.2022.07.060
– ident: ref2
  doi: 10.1609/aaai.v29i1.9491
– ident: ref15
  doi: 10.1109/MSP.2017.2693418
– ident: ref10
  doi: 10.1016/j.inffus.2023.101909
– ident: ref4
  doi: 10.1145/2939672.2939754
– ident: ref18
  doi: 10.1109/tbdata.2021.3081431
– ident: ref38
  doi: 10.1109/CVPR.2015.7298682
– ident: ref37
  doi: 10.3390/rs9030243
– year: 2018
  ident: ref41
  article-title: UMAP: Uniform manifold approximation and projection for dimension reduction
  publication-title: arXiv:1802.03426
– ident: ref28
  doi: 10.1109/TGRS.2020.3016020
– ident: ref12
  doi: 10.3390/rs14194872
– ident: ref27
  doi: 10.18653/v1/2020.acl-main.617
– ident: ref29
  doi: 10.1109/TGRS.2021.3073269
– ident: ref30
  doi: 10.3390/rs14051161
– volume: 1050
  start-page: 48550
  issue: 20
  year: 2017
  ident: ref6
  article-title: Graph attention networks
  publication-title: Stat
– ident: ref8
  doi: 10.1016/j.isprsjprs.2021.03.016
– ident: ref33
  doi: 10.3390/rs13214201
SSID ssj0014517
Score 2.452944
Snippet The recognition of complex geographic scenes requires to simultaneously describe different kinds of relations such as hyponymy and spatial relations among...
SourceID proquest
crossref
ieee
SourceType Aggregation Database
Index Database
Publisher
StartPage 1
SubjectTerms Artificial neural networks
Computational modeling
Embedding
Euclidean geometry
Euclidean space
Feature extraction
Geographic scenes
Geometric transformation
Graph representations
Graphical representations
Hyperbolic coordinates
hyperbolic space
Knowledge graphs
Knowledge representation
Machine learning
multirelations
Neural networks
Object recognition
Quaternions
Remote sensing
Semantic segmentation
Semantics
Vectors
Visualization
Title Representation of Multirelations of Geographic Scenes Based on Hyperbolic Space
URI https://ieeexplore.ieee.org/document/10543074
https://www.proquest.com/docview/3066914745
Volume 62
WOSCitedRecordID wos001246585400047&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/IET Electronic Library (IEL) (UW System Shared)
  customDbUrl:
  eissn: 1558-0644
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0014517
  issn: 0196-2892
  databaseCode: RIE
  dateStart: 19800101
  isFulltext: true
  titleUrlDefault: https://ieeexplore.ieee.org/
  providerName: IEEE
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV07T8MwED7RCiQYeJQiCgVlYEJKiRM7iUdAlE4FtUXqFsUv0aVFacvv5-y4qAgxsFmRnUR3vjt_vhfATZyWOmGShZKJLKRZYkIexzpUkYwUkSiaUrhmE9lwmE-n_NUnq7tcGK21Cz7TPTt0vny1kGt7VYYSzijuSdqARpaldbLWt8uAMuJzo9MQUUTsXZgk4neT59EYoWBMe4l1W9HkhxFyXVV-qWJnX_pH__yzYzj0B8ngvub8CezoeQsOtsoLtmDPhXfK5Sm8jFzAq88zmgcLE7jM22oTCmef-H7o7zMZjKVVgcEDmjgV4PwBotVK2BLCwRhBtm7DW_9p8jgIfSuFUMY0XYV5aRLB0R4aLvM8MSLDsSC5ULFhac5zqWOFZzNuDCMlopayRJxHjDQRKiChkjNozhdzfQ6BIbKkHIEVWn5qCM8RsSlLdUEkF4p14HZD2-KjrphROKQR8cIyorCMKDwjOtC2xNyaWNOxA90NOwovVMsC0U2KH84ou_hj2SXs27fXVyRdaK6qtb6CXfm5mi2ra7dfvgBPqL4S
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV07T8MwED7xFDDwLKJQIAMTUiB27CQeAVGKgIJokdii-CVYWtQWfj9nx0UgxMBmRbYc3fnu_PleAEc0q0zKFY8Vl3nM8tTGglIT60QlmigUTSV9s4m82y2en8VDSFb3uTDGGB98Zk7c0Pvy9VC9u6cylHDO8EyyWZjnjNGkTtf6chowTkJ2dBYjjqDBiUkScdq_euwhGKTsJHWOK5b-MEO-r8ovZewtTHvtn_-2DqvhKhmd1bzfgBkz2ISVbwUGN2HRB3iq8RbcP_qQ15BpNIiGNvK5t6NpMJz7Ejqiv7yqqKecEozO0cjpCOd3EK-OpCsiHPUQZpsGPLUv-xedODRTiBVl2SQuKptKgRbRClUUqZU5jiUppKaWZ4UolKEab2fCWk4qxC1VhUiPWGUTVEFSp9swNxgOzA5ElqiKCYRWaPuZJaJAzKYd1SVRQmrehOMpbcu3umZG6bFGIkrHiNIxogyMaELDEfPbxJqOTWhN2VEGsRqXiG8y3DhnfPePZYew1Onf3Za3192bPVh2O9UPJi2Ym4zezT4sqI_J63h04M_OJ1-OwVk
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=Representation+of+Multirelations+of+Geographic+Scenes+Based+on+Hyperbolic+Space&rft.jtitle=IEEE+transactions+on+geoscience+and+remote+sensing&rft.au=Cui%2C+Wei&rft.au=Xu%2C+Xing&rft.au=Chen%2C+Jiale&rft.au=Feng%2C+Zhanyun&rft.date=2024&rft.pub=IEEE&rft.issn=0196-2892&rft.volume=62&rft.spage=1&rft.epage=13&rft_id=info:doi/10.1109%2FTGRS.2024.3407843&rft.externalDocID=10543074
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0196-2892&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0196-2892&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0196-2892&client=summon