Indoor Instance-Aware Semantic Mapping Using Instance Segmentation

In order to accomplish the requirement of scene understanding to complete various kinds of complex tasks in home environment for robots, a novel instance segmentation method is adopted to build an instance-level 3D semantic map and obtain information such as categories, positions and interrelationsh...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Chinese Control and Decision Conference s. 3549 - 3554
Hlavní autoři: Jiang, Yinpeng, Ma, Xudong, Fang, Fang, Kang, Xuewen
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 22.05.2021
Témata:
ISSN:1948-9447
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Abstract In order to accomplish the requirement of scene understanding to complete various kinds of complex tasks in home environment for robots, a novel instance segmentation method is adopted to build an instance-level 3D semantic map and obtain information such as categories, positions and interrelationship of instance objects within the environment. Different from the previous method which focuses on a certain feature in geometry or vision, we synchronously learn the features of geometric and visual information, distinguish instance objects and background areas and create the feature voxel grid of the environment. The proposed 3D-RPN network takes the grid as input and makes use of the cuboid bounding box to predict each instance and the category it represents. With the mask prediction branch, we binarized voxels in each bounding box to determine the exact distribution of the instance object. Our method borrows the idea of Mask R-CNN and the main body is constructed by 3D and 2D convolutional network, making full use of the features of 2D and 3D. We have tested our method on ScanNet and S3DIS, two large-scale indoor scene data sets, and the experiment has verified that our method can find and identify the instance information more accurately than previous methods.
AbstractList In order to accomplish the requirement of scene understanding to complete various kinds of complex tasks in home environment for robots, a novel instance segmentation method is adopted to build an instance-level 3D semantic map and obtain information such as categories, positions and interrelationship of instance objects within the environment. Different from the previous method which focuses on a certain feature in geometry or vision, we synchronously learn the features of geometric and visual information, distinguish instance objects and background areas and create the feature voxel grid of the environment. The proposed 3D-RPN network takes the grid as input and makes use of the cuboid bounding box to predict each instance and the category it represents. With the mask prediction branch, we binarized voxels in each bounding box to determine the exact distribution of the instance object. Our method borrows the idea of Mask R-CNN and the main body is constructed by 3D and 2D convolutional network, making full use of the features of 2D and 3D. We have tested our method on ScanNet and S3DIS, two large-scale indoor scene data sets, and the experiment has verified that our method can find and identify the instance information more accurately than previous methods.
Author Jiang, Yinpeng
Ma, Xudong
Fang, Fang
Kang, Xuewen
Author_xml – sequence: 1
  givenname: Yinpeng
  surname: Jiang
  fullname: Jiang, Yinpeng
  email: jyp_seu@seu.edu.cn
  organization: Academy of Automation, Southeast University,Nanjing,210000
– sequence: 2
  givenname: Xudong
  surname: Ma
  fullname: Ma, Xudong
  organization: Academy of Automation, Southeast University,Nanjing,210000
– sequence: 3
  givenname: Fang
  surname: Fang
  fullname: Fang, Fang
  organization: Academy of Automation, Southeast University,Nanjing,210000
– sequence: 4
  givenname: Xuewen
  surname: Kang
  fullname: Kang, Xuewen
  email: 530100265@qq.com
  organization: Huaibei Normal University,Huaibei,China,235000
BookMark eNo1j0FLw0AUhFdRsK39BYLkD6Tue293s3ussWqg4kF7LpvkpayYTUgC4r-3Yr3MHOZjmJmLi9hFFuIW5ApAurs8f8g1EuAKJcLKGYlo8UwsXWbBGK2UtM6dixk4ZVOnVHYl5uP4IaUxJOVM3Bex7rohKeI4-Vhxuv7yAydv3Po4hSp58X0f4iHZjb_6Tx3zQ8tx8lPo4rW4bPznyMuTL8TucfOeP6fb16ciX2_TgJKmFLB0NtNgm-PIqq40U2aZGlIEVMustqRKgwYZgcE1rFWlsSZSytclEC3EzV9vYOZ9P4TWD9_702X6ASKqS_4
ContentType Conference Proceeding
DBID 6IE
6IL
CBEJK
RIE
RIL
DOI 10.1109/CCDC52312.2021.9602282
DatabaseName IEEE Electronic Library (IEL) Conference Proceedings
IEEE Proceedings Order Plan All Online (POP All Online) 1998-present by volume
IEEE Xplore All Conference Proceedings
IEEE/IET Electronic Library (IEL) (UW System Shared)
IEEE Proceedings Order Plans (POP All) 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 Engineering
EISBN 9781665440899
1665440899
EISSN 1948-9447
EndPage 3554
ExternalDocumentID 9602282
Genre orig-research
GrantInformation_xml – fundername: National Nature Science Foundation
  grantid: 61573100
  funderid: 10.13039/501100001809
GroupedDBID 29B
6IE
6IF
6IK
6IL
6IN
AAJGR
AAWTH
ABLEC
ADZIZ
ALMA_UNASSIGNED_HOLDINGS
BEFXN
BFFAM
BGNUA
BKEBE
BPEOZ
CBEJK
CHZPO
IEGSK
IPLJI
M43
OCL
RIE
RIL
ID FETCH-LOGICAL-i203t-12b987518f602cdc5e378e3f34313d07d834b6262e21e19fe54c52d3344adb133
IEDL.DBID RIE
ISICitedReferencesCount 2
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000824370103106&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 04:58:05 EDT 2025
IsPeerReviewed false
IsScholarly false
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-i203t-12b987518f602cdc5e378e3f34313d07d834b6262e21e19fe54c52d3344adb133
PageCount 6
ParticipantIDs ieee_primary_9602282
PublicationCentury 2000
PublicationDate 2021-May-22
PublicationDateYYYYMMDD 2021-05-22
PublicationDate_xml – month: 05
  year: 2021
  text: 2021-May-22
  day: 22
PublicationDecade 2020
PublicationTitle Chinese Control and Decision Conference
PublicationTitleAbbrev CCDC
PublicationYear 2021
Publisher IEEE
Publisher_xml – name: IEEE
SSID ssj0066300
Score 1.7728525
Snippet In order to accomplish the requirement of scene understanding to complete various kinds of complex tasks in home environment for robots, a novel instance...
SourceID ieee
SourceType Publisher
StartPage 3549
SubjectTerms 3D Semantic Map
3D-RPN
Feature extraction
Feature Voxel Grid
Geometry
Instance Segmentation
Mask Feature
Navigation
Semantics
Solid modeling
Three-dimensional displays
Visualization
Title Indoor Instance-Aware Semantic Mapping Using Instance Segmentation
URI https://ieeexplore.ieee.org/document/9602282
WOSCitedRecordID wos000824370103106&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/eLvHCXMwlV07T8MwED61FQMsPFrEWxkYSRu_kniEQEUlqCoBUrcqti9VhyYopPD3sdNQQGJhs_yQrbPlu8_n7w7gMuaKorA7QFKlfG7PkB-nMvKFyUKqJUuzrM5a8hCNx_F0KictuNpwYRCx_nyGfVesffmm0Cv3VDaw1ja1EKEN7SgK11ytr1s3dKGjGgYwCeQgSW4Ti7GI41pR0m9G_kqhUmuQ4e7_5t6D3jcVz5tslMw-tDA_gJ0fUQS7cDPKTVGU3qi29DT61x9pid4TLq3UFtp7TF0MhrlX_w7Y9LLt82VDPMp78DK8e07u_SY1gr-gAat8QpWMnccks8vSRgtkUYwsY9YeYCaITMy4sliFIiVIZIaCa0ENY5ynRllcegidvMjxCDyqOdEqpZFFijzMtFIiMEpKbgRa3SWOoeukMXtdR7-YNYI4-bv6FLadwJ1_ndIz6FTlCs9hS79Xi7fyot6yTwa6lpY
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LTwIxEJ4gmqgXH2h8uwePLmxf7PaoKIEIhERMuJFtO0s4sGtW0L9vu6yoiRdvTR9pM20683X6zQDcRFxRFHYHSKyUz-0Z8qNYhr4wSZNqyeIkKbKW9MLBIBqP5bACt2suDCIWn8-w7oqFL99keumeyhrW2qYWImzApuCcBiu21te923TBo0oOMAlko9V6aFmURRzbipJ6OfZXEpVCh7T3_jf7Phx9k_G84VrNHEAF00PY_RFHsAb33dRkWe51C1tPo3_3EefoPePcym2mvX7sojBMveJ_wLqXbZ_OS-pRegQv7cdRq-OXyRH8GQ3YwidUycj5TBK7LG20QBZGyBJmLQJmgtBEjCuLVihSgkQmKLgW1DDGeWyURabHUE2zFE_Ao5oTrWIaWqzIm4lWSgRGScmNQKu9xCnUnDQmr6v4F5NSEGd_V1_DdmfU70163cHTOew44TtvO6UXUF3kS7yELf2-mL3lV8X2fQIOi5nd
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=Chinese+Control+and+Decision+Conference&rft.atitle=Indoor+Instance-Aware+Semantic+Mapping+Using+Instance+Segmentation&rft.au=Jiang%2C+Yinpeng&rft.au=Ma%2C+Xudong&rft.au=Fang%2C+Fang&rft.au=Kang%2C+Xuewen&rft.date=2021-05-22&rft.pub=IEEE&rft.eissn=1948-9447&rft.spage=3549&rft.epage=3554&rft_id=info:doi/10.1109%2FCCDC52312.2021.9602282&rft.externalDocID=9602282