D3Feat: Joint Learning of Dense Detection and Description of 3D Local Features

A successful point cloud registration often lies on robust establishment of sparse matches through discriminative 3D local features. Despite the fast evolution of learning-based 3D feature descriptors, little attention has been drawn to the learning of 3D feature detectors, even less for a joint lea...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Proceedings (IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Online) s. 6358 - 6366
Hlavní autoři: Bai, Xuyang, Luo, Zixin, Zhou, Lei, Fu, Hongbo, Quan, Long, Tai, Chiew-Lan
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 01.06.2020
Témata:
ISSN:1063-6919
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Abstract A successful point cloud registration often lies on robust establishment of sparse matches through discriminative 3D local features. Despite the fast evolution of learning-based 3D feature descriptors, little attention has been drawn to the learning of 3D feature detectors, even less for a joint learning of the two tasks. In this paper, we leverage a 3D fully convolutional network for 3D point clouds, and propose a novel and practical learning mechanism that densely predicts both a detection score and a description feature for each 3D point. In particular, we propose a keypoint selection strategy that overcomes the inherent density variations of 3D point clouds, and further propose a self-supervised detector loss guided by the on-the-fly feature matching results during training. Finally, our method achieves state-of-the-art results in both indoor and outdoor scenarios, evaluated on 3DMatch and KITTI datasets, and shows its strong generalization ability on the ETH dataset. Towards practical use, we show that by adopting a reliable feature detector, sampling a smaller number of features is sufficient to achieve accurate and fast point cloud alignment.
AbstractList A successful point cloud registration often lies on robust establishment of sparse matches through discriminative 3D local features. Despite the fast evolution of learning-based 3D feature descriptors, little attention has been drawn to the learning of 3D feature detectors, even less for a joint learning of the two tasks. In this paper, we leverage a 3D fully convolutional network for 3D point clouds, and propose a novel and practical learning mechanism that densely predicts both a detection score and a description feature for each 3D point. In particular, we propose a keypoint selection strategy that overcomes the inherent density variations of 3D point clouds, and further propose a self-supervised detector loss guided by the on-the-fly feature matching results during training. Finally, our method achieves state-of-the-art results in both indoor and outdoor scenarios, evaluated on 3DMatch and KITTI datasets, and shows its strong generalization ability on the ETH dataset. Towards practical use, we show that by adopting a reliable feature detector, sampling a smaller number of features is sufficient to achieve accurate and fast point cloud alignment.
Author Tai, Chiew-Lan
Fu, Hongbo
Luo, Zixin
Bai, Xuyang
Quan, Long
Zhou, Lei
Author_xml – sequence: 1
  givenname: Xuyang
  surname: Bai
  fullname: Bai, Xuyang
  organization: Hong Kong University of Science and Technology
– sequence: 2
  givenname: Zixin
  surname: Luo
  fullname: Luo, Zixin
  organization: Hong Kong University of Science and Technology
– sequence: 3
  givenname: Lei
  surname: Zhou
  fullname: Zhou, Lei
  organization: Hong Kong University of Science and Technology
– sequence: 4
  givenname: Hongbo
  surname: Fu
  fullname: Fu, Hongbo
  organization: City University of Hong Kong
– sequence: 5
  givenname: Long
  surname: Quan
  fullname: Quan, Long
  organization: Hong Kong University of Science and Technology
– sequence: 6
  givenname: Chiew-Lan
  surname: Tai
  fullname: Tai, Chiew-Lan
  organization: Hong Kong University of Science and Technology
BookMark eNotjM1OwzAQhA0CiVLyBHDwCyTs2o6T5YZSyo8iQAi4Vo6zQUbFqZJw4O0JP4eZ0TcazbE4iH1kIc4QMkSg8-r18ckoC5ApUJABWE17IqGixELNQlvm-2KBc59aQjoSyTi-A4BWiJbKhbhf6TW76ULe9SFOsmY3xBDfZN_JFceRZ5_YT6GP0sV2ptEPYffL80SvZN17t5U_H58DjyfisHPbkZP_XIqX9dVzdZPWD9e31WWdBgV6ShGb1ncdOoLGWCg9W2PJOGy9Idam6VrlLBWkmtw3WOa6QdO1WntHPi-dXorTv9_AzJvdED7c8LUhzAurjP4G_ttQYA
CODEN IEEPAD
ContentType Conference Proceeding
DBID 6IE
6IH
CBEJK
RIE
RIO
DOI 10.1109/CVPR42600.2020.00639
DatabaseName IEEE Electronic Library (IEL) Conference Proceedings
IEEE Proceedings Order Plan (POP) 1998-present by volume
IEEE Xplore All Conference Proceedings
IEEE Electronic Library (IEL)
IEEE Proceedings Order Plans (POP) 1998-present
DatabaseTitleList
Database_xml – sequence: 1
  dbid: RIE
  name: IEEE Electronic Library (IEL)
  url: https://ieeexplore.ieee.org/
  sourceTypes: Publisher
DeliveryMethod fulltext_linktorsrc
Discipline Applied Sciences
EISBN 9781728171685
1728171687
EISSN 1063-6919
EndPage 6366
ExternalDocumentID 9157624
Genre orig-research
GroupedDBID 6IE
6IH
6IL
6IN
AAWTH
ABLEC
ADZIZ
ALMA_UNASSIGNED_HOLDINGS
BEFXN
BFFAM
BGNUA
BKEBE
BPEOZ
CBEJK
CHZPO
IEGSK
IJVOP
OCL
RIE
RIL
RIO
ID FETCH-LOGICAL-i203t-11bdcff1a90b4608ce64694a1dc49e34bfd2a69792b5cb1853b14fd33ca9c58a3
IEDL.DBID RIE
ISICitedReferencesCount 372
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000620679506063&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
IngestDate Wed Aug 27 02:30:35 EDT 2025
IsPeerReviewed false
IsScholarly true
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-i203t-11bdcff1a90b4608ce64694a1dc49e34bfd2a69792b5cb1853b14fd33ca9c58a3
PageCount 9
ParticipantIDs ieee_primary_9157624
PublicationCentury 2000
PublicationDate 2020-Jun
PublicationDateYYYYMMDD 2020-06-01
PublicationDate_xml – month: 06
  year: 2020
  text: 2020-Jun
PublicationDecade 2020
PublicationTitle Proceedings (IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Online)
PublicationTitleAbbrev CVPR
PublicationYear 2020
Publisher IEEE
Publisher_xml – name: IEEE
SSID ssj0003211698
Score 2.6265204
Snippet A successful point cloud registration often lies on robust establishment of sparse matches through discriminative 3D local features. Despite the fast evolution...
SourceID ieee
SourceType Publisher
StartPage 6358
SubjectTerms Convolution
Detectors
Feature extraction
Kernel
Three-dimensional displays
Training
Two dimensional displays
Title D3Feat: Joint Learning of Dense Detection and Description of 3D Local Features
URI https://ieeexplore.ieee.org/document/9157624
WOSCitedRecordID wos000620679506063&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
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LSwMxEB5q8eCpaiu-ycGjazc72Ue8thaRUopo6a1ksxPpZVfarb_fJF1WD168hCSEBGZIZr7JPADurA4em0xgQFLHgUh0EcjYRAE3mGmlKdQ-z_Zims5m2XIp5x24b2NhiMg7n9GD6_q__KLSO2cqG0puteNIHMBBmib7WK3WnoIWySQya6LjeCiHo8X81edftygw8g5criL4rxoqXoRMev87_BgGP7F4bN5KmRPoUHkKvUZ5ZM3V3PZhNkanzj2yl2pd1qzJm_rBKsPGFqqSbWvvdlUyVRZ21L4XbgmO2dRJNeb22FkIPoD3ydPb6DloiiUE6yjEOuA8L7QxXMkwF0mYaUos8hWKF1pIQpGbIlKJTGWUxzp3UjrnwhSIWlkuZQrPoFtWJZ0Di50xiDDXPEQLv6TKNKZkIThhhGkcX0DfkWf1uc-HsWooc_n39BUcOfrv3auuoVtvdnQDh_qrXm83t56J303dnCM
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV07T8MwED6VggRTgRbxxgMjofEriVlbqgIhqlCpulWO46AuCWpTfj-2GwUGFhbLtixbupN9953vAXBrdHCeR4x6WijusUBlnuA58XBOIyWV9pXLsz2LwySJ5nMxacFdEwujtXbOZ_redt1fflaqjTWV9QU22jFhO7DLGSP-NlqrsahQg2UCEdXxcdgX_cFs8uYysBscSJwLl60J_quKihMio87_jj-E3k80Hpo0cuYIWro4hk6tPqL6cq67kAypVege0HO5LCpUZ079QGWOhgasatNWzvGqQLLIzKh5MewSOkSxlWvI7rExILwH76PH6WDs1eUSvCXxaeVhnGYqz7EUfsoCP1I6MNiXSZwpJjRlaZ4RGYhQkJSr1MrpFLM8o1RJw6dI0hNoF2WhTwFxaw7SNFXYpwaACRkpGmoDwjUlNOT8DLqWPIvPbUaMRU2Z87-nb2B_PH2NF_FT8nIBB5YXW2erS2hXq42-gj31VS3Xq2vH0G-fBp9q
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=Proceedings+%28IEEE+Computer+Society+Conference+on+Computer+Vision+and+Pattern+Recognition.+Online%29&rft.atitle=D3Feat%3A+Joint+Learning+of+Dense+Detection+and+Description+of+3D+Local+Features&rft.au=Bai%2C+Xuyang&rft.au=Luo%2C+Zixin&rft.au=Zhou%2C+Lei&rft.au=Fu%2C+Hongbo&rft.date=2020-06-01&rft.pub=IEEE&rft.eissn=1063-6919&rft.spage=6358&rft.epage=6366&rft_id=info:doi/10.1109%2FCVPR42600.2020.00639&rft.externalDocID=9157624