Multi Queue for Unsupervised Person Re-identification

Recently, cluster-based methods have achieved significant success in unsupervised re-ID tasks. The hierarchical clustering algorithm, exemplified by SpCL, has been widely adopted in unsupervised cross-domain adaptation and unsupervised learning. The momentum-based feature update mechanism in SpCL ha...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Proceedings of the ... IEEE International Conference on Acoustics, Speech and Signal Processing (1998) s. 1 - 5
Hlavní autoři: Lin, Zhenyuan, Xie, Shengyong, Liu, Danhua, Li, Weikun, Gao, Ang, Dong, Yubo
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 06.04.2025
Témata:
ISSN:2379-190X
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Abstract Recently, cluster-based methods have achieved significant success in unsupervised re-ID tasks. The hierarchical clustering algorithm, exemplified by SpCL, has been widely adopted in unsupervised cross-domain adaptation and unsupervised learning. The momentum-based feature update mechanism in SpCL has been integrated into various algorithms, achieving notable results in subsequent studies. In this paper, we propose a multi-queue feature updating algorithm that stores feature vectors corresponding to person IDs in multiple queues. Random sampling is then applied to construct the negative sample matrix for contrastive loss, addressing the limitations of momentum-based updating methods. Additionally, we replace the static temperature coefficient in contrastive loss with a trainable temperature coefficient, enabling the model to automatically balance sensitivity between easy and hard samples. The code is available at https://github.com/bmfarer/multi-queue.git.
AbstractList Recently, cluster-based methods have achieved significant success in unsupervised re-ID tasks. The hierarchical clustering algorithm, exemplified by SpCL, has been widely adopted in unsupervised cross-domain adaptation and unsupervised learning. The momentum-based feature update mechanism in SpCL has been integrated into various algorithms, achieving notable results in subsequent studies. In this paper, we propose a multi-queue feature updating algorithm that stores feature vectors corresponding to person IDs in multiple queues. Random sampling is then applied to construct the negative sample matrix for contrastive loss, addressing the limitations of momentum-based updating methods. Additionally, we replace the static temperature coefficient in contrastive loss with a trainable temperature coefficient, enabling the model to automatically balance sensitivity between easy and hard samples. The code is available at https://github.com/bmfarer/multi-queue.git.
Author Lin, Zhenyuan
Dong, Yubo
Li, Weikun
Gao, Ang
Liu, Danhua
Xie, Shengyong
Author_xml – sequence: 1
  givenname: Zhenyuan
  surname: Lin
  fullname: Lin, Zhenyuan
  email: linzhenyuan@stu.xidian.edu.cn
  organization: Xidian University,Guangzhou Institute of Technology,Guangzhou,China
– sequence: 2
  givenname: Shengyong
  surname: Xie
  fullname: Xie, Shengyong
  email: 2416057194@qq.com
  organization: Guilin University Of Electronic Technology,School of Computer Science and Information Security,Guilin,China
– sequence: 3
  givenname: Danhua
  surname: Liu
  fullname: Liu, Danhua
  email: dhliu@xidian.edu.cn
  organization: Xidian University,School of Artificial Intelligence,Xian,China
– sequence: 4
  givenname: Weikun
  surname: Li
  fullname: Li, Weikun
  email: liweikun1105@163.com
  organization: Guilin University Of Electronic Technology,School of Computer Science and Information Security,Guilin,China
– sequence: 5
  givenname: Ang
  surname: Gao
  fullname: Gao, Ang
  email: anggao@stu.xidian.edu.cn
  organization: Xidian University,School of Artificial Intelligence,Xian,China
– sequence: 6
  givenname: Yubo
  surname: Dong
  fullname: Dong, Yubo
  email: ybdong@stu.xidian.edu.cn
  organization: Xidian University,School of Artificial Intelligence,Xian,China
BookMark eNo1j1FLwzAUhaMouM39Ax_qD-i8Sdr03kcZOoWJ003wbaTJLURmOppW8N9bUJ8Oh8N3OGcqzmIbWYhrCQspgW4el7fb7aYgY2ChQJULCYhklDoRc6oIdQnaYFXIUzFRuqJcErxfiGlKHwAwBjgR5dNw6EP2MvDAWdN22VtMw5G7r5DYZxvuUhuzV86D59iHJjjbhzZeivPGHhLP_3Qmdvd3u-VDvn5ejavWeSDd50qBri2iw4JqZlYM1DBr46k2DqxBZx0XhLUefaWaChmkd6Um45UnPRNXv7VhpPfHLnza7nv__1L_AHhhSj0
ContentType Conference Proceeding
DBID 6IE
6IH
CBEJK
RIE
RIO
DOI 10.1109/ICASSP49660.2025.10889622
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 Engineering
EISBN 9798350368741
EISSN 2379-190X
EndPage 5
ExternalDocumentID 10889622
Genre orig-research
GroupedDBID 23M
6IE
6IF
6IH
6IK
6IL
6IM
6IN
AAJGR
AAWTH
ABLEC
ACGFS
ADZIZ
ALMA_UNASSIGNED_HOLDINGS
BEFXN
BFFAM
BGNUA
BKEBE
BPEOZ
CBEJK
CHZPO
IEGSK
IJVOP
IPLJI
M43
OCL
RIE
RIL
RIO
RNS
ID FETCH-LOGICAL-i93t-2203ba88c849beee2e09fee36d9b6c0a68cace498b3b6c72f78e01dc5396d2d93
IEDL.DBID RIE
IngestDate Wed Nov 19 08:26:48 EST 2025
IsPeerReviewed false
IsScholarly true
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-i93t-2203ba88c849beee2e09fee36d9b6c0a68cace498b3b6c72f78e01dc5396d2d93
PageCount 5
ParticipantIDs ieee_primary_10889622
PublicationCentury 2000
PublicationDate 2025-April-6
PublicationDateYYYYMMDD 2025-04-06
PublicationDate_xml – month: 04
  year: 2025
  text: 2025-April-6
  day: 06
PublicationDecade 2020
PublicationTitle Proceedings of the ... IEEE International Conference on Acoustics, Speech and Signal Processing (1998)
PublicationTitleAbbrev ICASSP
PublicationYear 2025
Publisher IEEE
Publisher_xml – name: IEEE
SSID ssj0008748
Score 2.2878249
Snippet Recently, cluster-based methods have achieved significant success in unsupervised re-ID tasks. The hierarchical clustering algorithm, exemplified by SpCL, has...
SourceID ieee
SourceType Publisher
StartPage 1
SubjectTerms Classification algorithms
Clustering algorithms
hierarchical clustering algorithm
Image classification
multi queue features updating algorithm
Sensitivity
Signal processing
Signal processing algorithms
Speech processing
Temperature sensors
Unsupervised learning
unsupervised re-ID tasks
updating feature vectors
Vectors
Title Multi Queue for Unsupervised Person Re-identification
URI https://ieeexplore.ieee.org/document/10889622
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LTwMhECbaGKMXXzX1GUy8bsVlF4ajaWz00qy2Jr01PGaTXrZN2_X3C3Rb9eBBTjxCgCEMzDDfDCH3kGmd51onoEUZVDcmUcy4hJs854ojooYYbEIOBjAeq6IBq0csjG-NxmfYDdn4l-9mtg6qMn_CAZRIPcfdlVKuwVpbtgsyg31y1zjRfHjtPQ2HRRacT3opMM27m86_wqjEW6R_9M_xj0n7G49Hi-1Nc0J2sDolhz9cCZ6RPCJp6VuNNVL_EKUf1bKeB0awREeL-K6m75hMXWMeFHekTUb951HvJWlCIiRTxVdJmjJuNICFTBk_vRSZKhG5cMoIy7QAqy1mCgz3ZZmWEpA9OuvJLlzqFD8nrWpWYYdQLyYoZcH38jIeaqZdydBCCSIkJy5IOxBgMl87vZhs1n75R_0VOQhkjkYt4pq0Vosab8ie_VxNl4vbuFVf9pmVsw
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LTwIxEG4MGh8XXxjfronXxaXtdqdHQyQQkayCCTfSbWcTLgsB1t9vWxbUgwd76iOTtNN0ptPON0PIA3Cl4lipEJTI3dNNFsooMyHL4phJhogKfLKJpN-H0UimFVjdY2HsqHc-w4ar-r98M9WleyqzJxxACmol7nbMOW2u4FobwQsJh11yX4XRfOy2ngaDlLvwk9YOpHFjTf4rkYrXI-3Df87giNS_EXlButE1x2QLixNy8COY4CmJPZY2eCuxxMBeRYOPYlHOnChYoAlSf7MO3jGcmMpByO9JnQzbz8NWJ6ySIoQTyZYhpRHLFIAGLjM7PYqRzBGZMDITOlICtNLIJWTMthOaJ4BR02jLeGGokeyM1IppgecksIaClBoslbXyUEXK5BFqyEG4YsQFqTsGjGersBfj9dov_-i_I3ud4Wtv3Ov2X67IvmO5d3ER16S2nJd4Q3b053KymN_6bfsC8ciY-g
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+of+the+...+IEEE+International+Conference+on+Acoustics%2C+Speech+and+Signal+Processing+%281998%29&rft.atitle=Multi+Queue+for+Unsupervised+Person+Re-identification&rft.au=Lin%2C+Zhenyuan&rft.au=Xie%2C+Shengyong&rft.au=Liu%2C+Danhua&rft.au=Li%2C+Weikun&rft.date=2025-04-06&rft.pub=IEEE&rft.eissn=2379-190X&rft.spage=1&rft.epage=5&rft_id=info:doi/10.1109%2FICASSP49660.2025.10889622&rft.externalDocID=10889622