Unsupervised dense matching method and system based on reconstruction mapping consistency
The invention belongs to the technical field of dense matching, and relates to an unsupervised dense matching method and system based on reconstruction mapping consistency. The method comprises: constructing a dense matching network, and learning mapping from an input image to a disparity map by usi...
Saved in:
Main Authors | , , , , , , , , , , , , , |
---|---|
Format | Patent |
Language | Chinese English |
Published |
10.12.2021
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Abstract | The invention belongs to the technical field of dense matching, and relates to an unsupervised dense matching method and system based on reconstruction mapping consistency. The method comprises: constructing a dense matching network, and learning mapping from an input image to a disparity map by using an unsupervised loss function as a target constraint function, wherein the unsupervised loss function comprises a reconstruction mapping consistency loss function, a smooth loss function and a left-right consistency loss function; collecting scene sample data, and dividing the scene sample data into a training sample and a test sample; pre-training the dense matching network by using the training sample, and performing test optimization on the pre-trained network by using the test sample; and performing dense matching of the target scene data by using the dense matching network after test optimization. Reconstruction mapping consistent loss is utilized, smooth and left-right consistent loss is combined to serve |
---|---|
AbstractList | The invention belongs to the technical field of dense matching, and relates to an unsupervised dense matching method and system based on reconstruction mapping consistency. The method comprises: constructing a dense matching network, and learning mapping from an input image to a disparity map by using an unsupervised loss function as a target constraint function, wherein the unsupervised loss function comprises a reconstruction mapping consistency loss function, a smooth loss function and a left-right consistency loss function; collecting scene sample data, and dividing the scene sample data into a training sample and a test sample; pre-training the dense matching network by using the training sample, and performing test optimization on the pre-trained network by using the test sample; and performing dense matching of the target scene data by using the dense matching network after test optimization. Reconstruction mapping consistent loss is utilized, smooth and left-right consistent loss is combined to serve |
Author | JIN FEI WANG SHUXIANG GUAN KAI WEI LINSU GAO XUEMEI WANG JIANFENG RUI JIE LI HUA MIAO YUZHE LIN YUZHUN LIU ZHI WANG FAN GUO HAOJUN LIU XIAO |
Author_xml | – fullname: LIN YUZHUN – fullname: RUI JIE – fullname: WEI LINSU – fullname: WANG JIANFENG – fullname: LI HUA – fullname: MIAO YUZHE – fullname: GUO HAOJUN – fullname: WANG FAN – fullname: GUAN KAI – fullname: WANG SHUXIANG – fullname: JIN FEI – fullname: LIU ZHI – fullname: GAO XUEMEI – fullname: LIU XIAO |
BookMark | eNqNy00KwjAUBOAsdOHfHZ4HEFq6UJdSFFeudOGqxGS0AfMS8lKht7cFD-BqmOGbuZpwYMzU_cbSRaSPE1iyYAF5nU3r-EUeuQ2WNFuSXjI8PfTIAlOCCSw5dSa7oXod4_gYRzdINv1STZ_6LVj9cqHWp-O1Pm8QQwOJ2oCRm_pSltV2V1T74lD9Y76l-T16 |
ContentType | Patent |
DBID | EVB |
DatabaseName | esp@cenet |
DatabaseTitleList | |
Database_xml | – sequence: 1 dbid: EVB name: esp@cenet url: http://worldwide.espacenet.com/singleLineSearch?locale=en_EP sourceTypes: Open Access Repository |
DeliveryMethod | fulltext_linktorsrc |
Discipline | Medicine Chemistry Sciences Physics |
DocumentTitleAlternate | 基于重构映射一致的无监督密集匹配方法及系统 |
ExternalDocumentID | CN113780390A |
GroupedDBID | EVB |
ID | FETCH-epo_espacenet_CN113780390A3 |
IEDL.DBID | EVB |
IngestDate | Fri Jul 19 14:27:59 EDT 2024 |
IsOpenAccess | true |
IsPeerReviewed | false |
IsScholarly | false |
Language | Chinese English |
LinkModel | DirectLink |
MergedId | FETCHMERGED-epo_espacenet_CN113780390A3 |
Notes | Application Number: CN202111008712 |
OpenAccessLink | https://worldwide.espacenet.com/publicationDetails/biblio?FT=D&date=20211210&DB=EPODOC&CC=CN&NR=113780390A |
ParticipantIDs | epo_espacenet_CN113780390A |
PublicationCentury | 2000 |
PublicationDate | 20211210 |
PublicationDateYYYYMMDD | 2021-12-10 |
PublicationDate_xml | – month: 12 year: 2021 text: 20211210 day: 10 |
PublicationDecade | 2020 |
PublicationYear | 2021 |
RelatedCompanies | INFORMATION ENGINEERING UNIVERSITY UNIT 61363 OF PLA |
RelatedCompanies_xml | – name: UNIT 61363 OF PLA – name: INFORMATION ENGINEERING UNIVERSITY |
Score | 3.5014336 |
Snippet | The invention belongs to the technical field of dense matching, and relates to an unsupervised dense matching method and system based on reconstruction mapping... |
SourceID | epo |
SourceType | Open Access Repository |
SubjectTerms | CALCULATING COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS COMPUTING COUNTING HANDLING RECORD CARRIERS PHYSICS PRESENTATION OF DATA RECOGNITION OF DATA RECORD CARRIERS |
Title | Unsupervised dense matching method and system based on reconstruction mapping consistency |
URI | https://worldwide.espacenet.com/publicationDetails/biblio?FT=D&date=20211210&DB=EPODOC&locale=&CC=CN&NR=113780390A |
hasFullText | 1 |
inHoldings | 1 |
isFullTextHit | |
isPrint | |
link | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwfV1LS8NAEB5qfd40KlofrCC5BZunySGI3SQUoWmRVuqpbJJN1UMSSETw1zu7fXnR64Qsm1lm5vs2s98C3JqJmzPX8TTuckOzEjvTmJmkWs4NkxvMc5JM7EMOYqc_sZ6m9rQFH6uzMFIn9EuKI2JEpRjvjczX1WYTK5C9lfVd8o6m8iEa-4G6ZMfIZpDCqEHPD0fDYEhVSn0aq_Gzr-vmvdtFgv-4BdsCRgud_fClJ06lVL9LSnQIOyMcrWiOoPX9psA-Xd28psDeYPnDW4Fd2aGZ1mhcRmF9DK-Tov6sRJDXPCOYOGpOEHjKrkiyuBKasCIjC5VmIgpVRsqCSPK7FozFV4Q2w5wIo1hsTLMncBOFY9rXcLaztWtmNN58mHkK7aIs-BkQ00Agl3oe8xDv5NxDmstsRwLCnFm6cQ6dv8fp_PfwAg6Em0VDh969hDbOmV9hWW6Sa-nPH9Bnk3g |
link.rule.ids | 230,309,786,891,25594,76904 |
linkProvider | European Patent Office |
linkToHtml | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwfV1bT4MwFD6Z8zLfFDU6bzUxvBEHbAweiHFlBHVji2FmPpECxcsDW8KMib_e0-7mi76epk17mnP52tOvANdmYufMthyN29zQmkkr05iZpFrODZMbzLGSTJxD9kMrGDUfxq1xBT6Wb2EkT-iXJEdEi0rR3mfSX0_Xh1ierK0sb5J3FE1u_cj11AU6RjSDEEb1Om53OPAGVKXUpaEaPrm6brbtBgL8uw3YbCMkFDz73eeOeJUy_R1S_D3YGuJoxWwfKt9vCtTo8uc1BXb6iwtvBbZlhWZaonBhheUBvIyK8nMqjLzkGUHHUXKCiaesiiTzL6EJKzIyZ2kmIlBlZFIQCX5XhLHYRXAzvBIhFJuNbvYQrvxuRAMNZxuvVBPTcL0w8wiqxaTgx0BMAxO51HGYg_lOzh2EuaxlyYQwZ03dOIH63-PU_2u8hFoQ9Xtx7z58PIVdoXJR3KE3zqCK8-fnGKJnyYXU7Q_dA5Zj |
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%3Apatent&rft.title=Unsupervised+dense+matching+method+and+system+based+on+reconstruction+mapping+consistency&rft.inventor=LIN+YUZHUN&rft.inventor=RUI+JIE&rft.inventor=WEI+LINSU&rft.inventor=WANG+JIANFENG&rft.inventor=LI+HUA&rft.inventor=MIAO+YUZHE&rft.inventor=GUO+HAOJUN&rft.inventor=WANG+FAN&rft.inventor=GUAN+KAI&rft.inventor=WANG+SHUXIANG&rft.inventor=JIN+FEI&rft.inventor=LIU+ZHI&rft.inventor=GAO+XUEMEI&rft.inventor=LIU+XIAO&rft.date=2021-12-10&rft.externalDBID=A&rft.externalDocID=CN113780390A |