RealNet: A Feature Selection Network with Realistic Synthetic Anomaly for Anomaly Detection

Self-supervised feature reconstruction methods have shown promising advances in industrial image anomaly de-tection and localization. Despite this progress, these meth-ods still face challenges in synthesizing realistic and di-verse anomaly samples, as well as addressing the feature redundancy and p...

Full description

Saved in:
Bibliographic Details
Published in:Proceedings (IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Online) pp. 16699 - 16708
Main Authors: Zhang, Ximiao, Xu, Min, Zhou, Xiuzhuang
Format: Conference Proceeding
Language:English
Published: IEEE 16.06.2024
Subjects:
anomaly detection
anomaly synthesis
Computer vision
Data models
Feature extraction
feature selection
Location awareness
Pattern recognition
Reconstruction algorithms
Redundancy
ISSN:1063-6919
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Abstract Self-supervised feature reconstruction methods have shown promising advances in industrial image anomaly de-tection and localization. Despite this progress, these meth-ods still face challenges in synthesizing realistic and di-verse anomaly samples, as well as addressing the feature redundancy and pre-training bias of pre-trained feature. In this work, we introduce RealNet, a feature reconstruction network with realistic synthetic anomaly and adaptive feature selection. It is incorporated with three key inno-vations: First, we propose Strength-controllable Diffusion Anomaly Synthesis (SDAS), a diffusion process-based syn-thesis strategy capable of generating samples with varying anomaly strengths that mimic the distribution of real anomalous samples. Second, we develop Anomaly-aware Features Selection (A FS), a method for selecting repre-sentative and discriminative pre-trained feature subsets to improve anomaly detection performance while controlling computational costs. Third, we introduce Reconstruction Residuals Selection (RRS), a strategy that adaptively selects discriminative residuals for comprehensive identification of anomalous regions across multiple levels of granularity. We assess RealNet onfour benchmark datasets, and our results demonstrate significant improvements in both Image AU-Rae and Pixel AUROC compared to the current state-of-the-art methods. The code, data, and models are available at https://github.com/cnulab/RealNet.
AbstractList Self-supervised feature reconstruction methods have shown promising advances in industrial image anomaly de-tection and localization. Despite this progress, these meth-ods still face challenges in synthesizing realistic and di-verse anomaly samples, as well as addressing the feature redundancy and pre-training bias of pre-trained feature. In this work, we introduce RealNet, a feature reconstruction network with realistic synthetic anomaly and adaptive feature selection. It is incorporated with three key inno-vations: First, we propose Strength-controllable Diffusion Anomaly Synthesis (SDAS), a diffusion process-based syn-thesis strategy capable of generating samples with varying anomaly strengths that mimic the distribution of real anomalous samples. Second, we develop Anomaly-aware Features Selection (A FS), a method for selecting repre-sentative and discriminative pre-trained feature subsets to improve anomaly detection performance while controlling computational costs. Third, we introduce Reconstruction Residuals Selection (RRS), a strategy that adaptively selects discriminative residuals for comprehensive identification of anomalous regions across multiple levels of granularity. We assess RealNet onfour benchmark datasets, and our results demonstrate significant improvements in both Image AU-Rae and Pixel AUROC compared to the current state-of-the-art methods. The code, data, and models are available at https://github.com/cnulab/RealNet.
Author Zhang, Ximiao
Xu, Min
Zhou, Xiuzhuang
Author_xml – sequence: 1
  givenname: Ximiao
  surname: Zhang
  fullname: Zhang, Ximiao
  email: 2211002048@cnu.edu.cn
  organization: College of Information and Engineering, Capital Normal University
– sequence: 2
  givenname: Min
  surname: Xu
  fullname: Xu, Min
  email: xumin@cnu.edu.cn
  organization: College of Information and Engineering, Capital Normal University
– sequence: 3
  givenname: Xiuzhuang
  surname: Zhou
  fullname: Zhou, Xiuzhuang
  email: xiuzhuang.zhou@bupt.edu.cn
  organization: School of Artificial Intelligence, Beijing University of Posts and Telecommunications
BookMark eNo9j81OAjEUhavRRETegEVfYLC3d9qZuiMoakLUgLpxQWp7G6rDjJkpIby9EIyr8yXnJzmX7KxuamJsCGIEIMz15P1lrmSBOJJC5iMBqhQnbGAKU6ISqFAIfcp6IDRm2oC5YIOu-xJCoATQpuyxjznZ6onSDR_zKdm0aYkvqCKXYlPzvbFt2m--jWnFD8nYpej4YlenFR1oXDdrW-14aNp_vqV0rF-x82CrjgZ_2mdv07vXyUM2e75_nIxnWYRCp8whFJKMd8F-Bo_CBaFLrXJdSB-MVTaH3Aefh1JpadF6UBIJfRGc9E4B9tnwuBuJaPnTxrVtd8v9aVUiAP4CPtRW9g
CODEN IEEPAD
ContentType Conference Proceeding
DBID 6IE
6IH
CBEJK
RIE
RIO
DOI 10.1109/CVPR52733.2024.01580
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 9798350353006
EISSN 1063-6919
EndPage 16708
ExternalDocumentID 10658311
Genre orig-research
GrantInformation_xml – fundername: National Natural Science Foundation of China
  grantid: 62177034,61972046
  funderid: 10.13039/501100001809
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-i176t-c3172e9dcfabfd30cf068654672df9a5a414dfd4f8562a3ad1523e3d7fc2dc513
IEDL.DBID RIE
ISICitedReferencesCount 74
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001342442408010&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 02:00:57 EDT 2025
IsPeerReviewed false
IsScholarly true
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-i176t-c3172e9dcfabfd30cf068654672df9a5a414dfd4f8562a3ad1523e3d7fc2dc513
PageCount 10
ParticipantIDs ieee_primary_10658311
PublicationCentury 2000
PublicationDate 2024-June-16
PublicationDateYYYYMMDD 2024-06-16
PublicationDate_xml – month: 06
  year: 2024
  text: 2024-June-16
  day: 16
PublicationDecade 2020
PublicationTitle Proceedings (IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Online)
PublicationTitleAbbrev CVPR
PublicationYear 2024
Publisher IEEE
Publisher_xml – name: IEEE
SSID ssj0003211698
Score 2.5999496
Snippet Self-supervised feature reconstruction methods have shown promising advances in industrial image anomaly de-tection and localization. Despite this progress,...
SourceID ieee
SourceType Publisher
StartPage 16699
SubjectTerms anomaly detection
anomaly synthesis
Computer vision
Data models
Feature extraction
feature selection
Location awareness
Pattern recognition
Reconstruction algorithms
Redundancy
Title RealNet: A Feature Selection Network with Realistic Synthetic Anomaly for Anomaly Detection
URI https://ieeexplore.ieee.org/document/10658311
WOSCitedRecordID wos001342442408010&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/eLvHCXMwlV3NS8MwFA86PHiaHxO_ycFrZ9t8Ld7GdHiQMjYdAw8jTV5goK1snbD_3iSt04sHL-VRUgIvpL9fXt7vPYRumIk9Lssot0pE1EgVSQ48ojln1GhNWejWMH0SWdabzeSoEasHLQwAhOQz6Hoz3OWbUq99qMztcIeXxCt5d4UQtVhrG1Ah7ijDZa-RxyWxvB1MR2NfX4y4Y2BKuw74fPHHX01UAoYM2_-c_QB1ftR4eLTFmUO0A8URajf0ETebc3WMXseO9GVQ3eE-9sxuvQQ8CW1unO9xVud7Yx94xX5kqNCMJ5vCUUBv9YvyXb1tsGOxW_seqvrzDnoZPjwPHqOmd0K0SASvIu14QQrSaKtya0isrdeCMPdbTI2ViimaUGMNtT1HgBRRxuE4AWKE1anRLCEnqFWUBZwibJmwORduqAEqgClrVSqV5nGau6c6Qx3vrPlHXR5j_u2n8z_eX6B9vx4-3yrhl6hVLddwhfb0Z7VYLa_Don4BeGekqw
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LSwMxEA5SBT3VR8W3OXjd2t08tvFWqqViXUpbS8FDyeYBBd1KuxX6751k1-rFg5dlWLIEJmS_L5P5ZhC6YbrhcFkEqZVxQLWQgeCGBzTljGqlKPPdGsa9OEmak4nol2J1r4UxxvjkM1N3pr_L13O1cqEy2OGAl8QpebcZpVFYyLU2IRUChxkumqVALmyI2_a4P3AVxggcBCNaB-hz5R9_tVHxKNKp_nP-fVT70ePh_gZpDtCWyQ5RtSSQuNyeyyP0OgDal5j8Drew43arhcFD3-gGvI-TIuMbu9ArdiN9jWY8XGdAAp3Vyubv8m2Ngcdu7HuTF5_X0EvnYdTuBmX3hGAWxjwPFDCDyAitrEytJg1lnRqEwY8x0lZIJmlItdXUNoECSSI1IDkxRMdWRVqxkByjSjbPzAnClsU25TEM1YbGhklrZSSk4o0ohac8RTXnrOlHUSBj-u2nsz_eX6Pd7ui5N-09Jk_naM-tjcu-CvkFquSLlblEO-ozny0XV36BvwA1T6fy
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=RealNet%3A+A+Feature+Selection+Network+with+Realistic+Synthetic+Anomaly+for+Anomaly+Detection&rft.au=Zhang%2C+Ximiao&rft.au=Xu%2C+Min&rft.au=Zhou%2C+Xiuzhuang&rft.date=2024-06-16&rft.pub=IEEE&rft.eissn=1063-6919&rft.spage=16699&rft.epage=16708&rft_id=info:doi/10.1109%2FCVPR52733.2024.01580&rft.externalDocID=10658311