Shuffle in 3D: A Lightweight Architecture for Stereo Matching

The deep learning-based stereo matching approaches commonly construct 3D cost volume with a Siamese network, and the 3D encoder-decoder architectures regularize 3D cost volume to generate disparity as output. Yet, the 3D encoder-decoder architectures produce tremendous parameters that causing huge t...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:2021 IEEE 10th Global Conference on Consumer Electronics (GCCE) s. 675 - 677
Hlavní autori: Xiao, Jianqiang, Yamane, Satoshi
Médium: Konferenčný príspevok..
Jazyk:English
Vydavateľské údaje: IEEE 12.10.2021
Predmet:
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Abstract The deep learning-based stereo matching approaches commonly construct 3D cost volume with a Siamese network, and the 3D encoder-decoder architectures regularize 3D cost volume to generate disparity as output. Yet, the 3D encoder-decoder architectures produce tremendous parameters that causing huge training and inference cost. In this paper, we propose a 3D ShuffleNet-based approach on PSMNet, which greatly saves the computational cost without losing without reducing the performance of the network.
AbstractList The deep learning-based stereo matching approaches commonly construct 3D cost volume with a Siamese network, and the 3D encoder-decoder architectures regularize 3D cost volume to generate disparity as output. Yet, the 3D encoder-decoder architectures produce tremendous parameters that causing huge training and inference cost. In this paper, we propose a 3D ShuffleNet-based approach on PSMNet, which greatly saves the computational cost without losing without reducing the performance of the network.
Author Yamane, Satoshi
Xiao, Jianqiang
Author_xml – sequence: 1
  givenname: Jianqiang
  surname: Xiao
  fullname: Xiao, Jianqiang
  email: kxiao@csl.ec.t.kanazawa-u.ac.jp
  organization: Kanazawa University,Institute of Electrical Engineering and Computer Science,Kanazawa,Japan
– sequence: 2
  givenname: Satoshi
  surname: Yamane
  fullname: Yamane, Satoshi
  email: syamane@is.t.kanazawa-u.ac.jp
  organization: Kanazawa University,Institute of Electrical Engineering and Computer Science,Kanazawa,Japan
BookMark eNotj9FKwzAYRiPohdt8AkHyAq1_kiZNBC9K3abQ4cUUvBtp8mcNzFZihvj2Ku7mOxcHDnwzcj5OIxJyw6BkDMztum2XUgDIkgNnpVGcgxZnZMaUkpVQtXq7JPfb4RjCAWkcqXi4ow3t4n7IX_i3tEluiBldPiakYUp0mzHhRDc2_4pxvyAXwR4-8erEOXldLV_ax6J7Xj-1TVdExnQudN2rYKz3WnqwUnruoMYQED1UDmTQynjea7DMMWNqX2mhJBouDO99MGJOrv-7ERF3Hym-2_S9Oz0SP8UNRVA
ContentType Conference Proceeding
DBID 6IE
6IL
CBEJK
RIE
RIL
DOI 10.1109/GCCE53005.2021.9622083
DatabaseName IEEE Electronic Library (IEL) Conference Proceedings
IEEE Xplore POP ALL
IEEE Xplore All Conference Proceedings
IEEE Electronic Library (IEL)
IEEE Proceedings Order Plans (POP All) 1998-Present
DatabaseTitleList
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
EISBN 166543676X
9781665436762
EndPage 677
ExternalDocumentID 9622083
Genre orig-research
GroupedDBID 6IE
6IL
CBEJK
RIE
RIL
ID FETCH-LOGICAL-i118t-87b6f9add85d0a55d2c07effeed04c05f869d2b80a1c1997d48365e92392bdf93
IEDL.DBID RIE
IngestDate Thu Jun 29 18:37:52 EDT 2023
IsPeerReviewed false
IsScholarly false
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-i118t-87b6f9add85d0a55d2c07effeed04c05f869d2b80a1c1997d48365e92392bdf93
PageCount 3
ParticipantIDs ieee_primary_9622083
PublicationCentury 2000
PublicationDate 2021-Oct.-12
PublicationDateYYYYMMDD 2021-10-12
PublicationDate_xml – month: 10
  year: 2021
  text: 2021-Oct.-12
  day: 12
PublicationDecade 2020
PublicationTitle 2021 IEEE 10th Global Conference on Consumer Electronics (GCCE)
PublicationTitleAbbrev GCCE
PublicationYear 2021
Publisher IEEE
Publisher_xml – name: IEEE
Score 1.7727991
Snippet The deep learning-based stereo matching approaches commonly construct 3D cost volume with a Siamese network, and the 3D encoder-decoder architectures...
SourceID ieee
SourceType Publisher
StartPage 675
SubjectTerms 3D ShuffleNet
Computational efficiency
Computer architecture
Conferences
Consumer electronics
Costs
lightweight 3D encoder-decoder
Stereo Matching
Three-dimensional displays
Training
Title Shuffle in 3D: A Lightweight Architecture for Stereo Matching
URI https://ieeexplore.ieee.org/document/9622083
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1JSwMxFA5t8eBJpRV3cvBo2kxmJpMIHkpt9VBLoQq9lSwvWJBpqa3-fZN0qApevIWQELK9970doWuuZRZwMvHPISeZdJRo_yGJomkKCXWp07FqybAYjcR0Ksc1dLOLhQGA6HwG7dCMtny7MJugKutIzpiHDHVULwq-jdWqgn4TKjsPvV4_D9nXvdTHknY1-FfVlMg0Bgf_W-4Qtb6j7_B4x1eOUA3KJrqbvG6cewM8L3F6f4u7eBjE6s-o2cTdH-YA7GEonvjjggV-8oQ2qJha6GXQf-49kqr0AZl7xL_2NEpzJz3tEbmlKs8tM7QIHh5gaWZo7gSXlmlBVWKCq4jNRMpz8GhNMm2dTI9Ro1yUcIIwFZAoLxcBZybjnAollZ9HRWaULiQ9Rc2w9dlym91iVu367O_uc7QfTpdEn44L1FivNnCJ9szHev6-uopX8gVc84yV
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV3NS8MwFH_MKehJZRO_zcGj3dI0SRPBw5ibE7sx2ITdRtMmOJBO5qb_vklWpoIXbyEkhHy993vfANdcSepwcmCfAwuoNDhQ9kMGKY4iHWITGeWrliTxYCAmEzmswM0mFkZr7Z3PdMM1vS0_n2crpyprSk6IhQxbsM0oJXgdrVWG_YZYNh_a7Q5z-det3EfCRjn8V90Uzza6-_9b8ADq3_F3aLjhLIdQ0UUN7kYvK2NeNZoVKLq_RS2UOMH60-s2UeuHQQBZIIpG9sD0HPUtqXVKpjo8dzvjdi8oix8EM4v5l5ZKKW6kpT6C5ThlLCcZjp2Ph84xzTAzgsucKIHTMHPOIjkVEWfa4jVJVG5kdATVYl7oY0BY6DC1kpHmJKOcY5HK1M7DgmapiiU-gZrb-vRtnd9iWu769O_uK9jtjfvJNHkcPJ3BnjvpwHt4nEN1uVjpC9jJPpaz98Wlv54v35uP3A
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%3Abook&rft.genre=proceeding&rft.title=2021+IEEE+10th+Global+Conference+on+Consumer+Electronics+%28GCCE%29&rft.atitle=Shuffle+in+3D%3A+A+Lightweight+Architecture+for+Stereo+Matching&rft.au=Xiao%2C+Jianqiang&rft.au=Yamane%2C+Satoshi&rft.date=2021-10-12&rft.pub=IEEE&rft.spage=675&rft.epage=677&rft_id=info:doi/10.1109%2FGCCE53005.2021.9622083&rft.externalDocID=9622083