Bridging the Gap Between Image Coding for Machines and Humans

Image coding for machines (ICM) aims at reducing the bitrate required to represent an image while minimizing the drop in machine vision analysis accuracy. In many use cases, such as surveillance, it is also important that the visual quality is not drastically deteriorated by the compression process....

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Proceedings - International Conference on Image Processing S. 3411 - 3415
Hauptverfasser: Le, Nam, Zhang, Honglei, Cricri, Francesco, Youvalari, Ramin G., Rezazadegan Tavakoli, Hamed, Aksu, Emre, Hannuksela, Miska M., Rahtu, Esa
Format: Tagungsbericht
Sprache:Englisch
Veröffentlicht: IEEE 16.10.2022
Schlagworte:
ISSN:2381-8549
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Abstract Image coding for machines (ICM) aims at reducing the bitrate required to represent an image while minimizing the drop in machine vision analysis accuracy. In many use cases, such as surveillance, it is also important that the visual quality is not drastically deteriorated by the compression process. Recent works on using neural network (NN) based ICM codecs have shown significant coding gains against traditional methods; however, the decompressed images, especially at low bitrates, often contain checkerboard artifacts. We propose an effective decoder finetuning scheme based on adversarial training to significantly enhance the visual quality of ICM codecs, while preserving the machine analysis accuracy, without adding extra bitcost or parameters at the inference phase. The results show complete removal of the checkerboard artifacts at the negligible cost of −1.6% relative change in task performance score. In the cases where some amount of artifacts is tolerable, such as when machine consumption is the primary target, this technique can enhance both pixel-fidelity and feature-fidelity scores without losing task performance.
AbstractList Image coding for machines (ICM) aims at reducing the bitrate required to represent an image while minimizing the drop in machine vision analysis accuracy. In many use cases, such as surveillance, it is also important that the visual quality is not drastically deteriorated by the compression process. Recent works on using neural network (NN) based ICM codecs have shown significant coding gains against traditional methods; however, the decompressed images, especially at low bitrates, often contain checkerboard artifacts. We propose an effective decoder finetuning scheme based on adversarial training to significantly enhance the visual quality of ICM codecs, while preserving the machine analysis accuracy, without adding extra bitcost or parameters at the inference phase. The results show complete removal of the checkerboard artifacts at the negligible cost of −1.6% relative change in task performance score. In the cases where some amount of artifacts is tolerable, such as when machine consumption is the primary target, this technique can enhance both pixel-fidelity and feature-fidelity scores without losing task performance.
Author Cricri, Francesco
Rahtu, Esa
Aksu, Emre
Le, Nam
Rezazadegan Tavakoli, Hamed
Zhang, Honglei
Youvalari, Ramin G.
Hannuksela, Miska M.
Author_xml – sequence: 1
  givenname: Nam
  surname: Le
  fullname: Le, Nam
  organization: Tampere University
– sequence: 2
  givenname: Honglei
  surname: Zhang
  fullname: Zhang, Honglei
  organization: Nokia Technologies
– sequence: 3
  givenname: Francesco
  surname: Cricri
  fullname: Cricri, Francesco
  organization: Nokia Technologies
– sequence: 4
  givenname: Ramin G.
  surname: Youvalari
  fullname: Youvalari, Ramin G.
  organization: Nokia Technologies
– sequence: 5
  givenname: Hamed
  surname: Rezazadegan Tavakoli
  fullname: Rezazadegan Tavakoli, Hamed
  organization: Nokia Technologies
– sequence: 6
  givenname: Emre
  surname: Aksu
  fullname: Aksu, Emre
  organization: Nokia Technologies
– sequence: 7
  givenname: Miska M.
  surname: Hannuksela
  fullname: Hannuksela, Miska M.
  organization: Nokia Technologies
– sequence: 8
  givenname: Esa
  surname: Rahtu
  fullname: Rahtu, Esa
  organization: Tampere University
BookMark eNotj8tKw0AYRkdRsKk-gSDzAon_zGQuWbiwQdtARRe6LnP5k46YSUki4tur2NXH4cCBLyNnaUhIyA2DgjGobpu6eSmV1KrgwHlRmUpXTJ2QjCkly0px0KdkwYVhufnlC5JN0zsABybYgtytxhi6mDo675Gu7YGucP5CTLTpbYe0HsKfbIeRPlm_jwknalOgm8_epumSnLf2Y8Kr4y7J2-PDa73Jt8_rpr7f5pGDmHN0Bq303qqytMZpZgNHr1uhJbRaclCaW1AycB7AK_AODJPOtaJynrVSLMn1fzci4u4wxt6O37vjVfEDhk5J0A
ContentType Conference Proceeding
DBID 6IE
6IH
CBEJK
RIE
RIO
DOI 10.1109/ICIP46576.2022.9897916
DatabaseName IEEE Electronic Library (IEL) Conference Proceedings
IEEE Proceedings Order Plan (POP) 1998-present by volume
IEEE Xplore All Conference Proceedings
IEEE/IET Electronic Library
IEEE Proceedings Order Plans (POP) 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
Discipline Applied Sciences
EISBN 1665496207
9781665496209
EISSN 2381-8549
EndPage 3415
ExternalDocumentID 9897916
Genre orig-research
GroupedDBID 6IE
6IF
6IH
6IK
6IL
6IM
6IN
AAJGR
AAWTH
ADZIZ
ALMA_UNASSIGNED_HOLDINGS
BEFXN
BFFAM
BGNUA
BKEBE
BPEOZ
CBEJK
CHZPO
IPLJI
M43
OCL
RIE
RIL
RIO
RNS
ID FETCH-LOGICAL-i203t-eb8ea5cca644a8b71ad2ec7f3750f7520672a065d22d0c60cb0815bbf39bc1f53
IEDL.DBID RIE
ISICitedReferencesCount 3
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001058109503101&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 02:19:00 EDT 2025
IsPeerReviewed false
IsScholarly true
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-i203t-eb8ea5cca644a8b71ad2ec7f3750f7520672a065d22d0c60cb0815bbf39bc1f53
PageCount 5
ParticipantIDs ieee_primary_9897916
PublicationCentury 2000
PublicationDate 2022-Oct.-16
PublicationDateYYYYMMDD 2022-10-16
PublicationDate_xml – month: 10
  year: 2022
  text: 2022-Oct.-16
  day: 16
PublicationDecade 2020
PublicationTitle Proceedings - International Conference on Image Processing
PublicationTitleAbbrev ICIP
PublicationYear 2022
Publisher IEEE
Publisher_xml – name: IEEE
SSID ssj0020131
Score 2.258474
Snippet Image coding for machines (ICM) aims at reducing the bitrate required to represent an image while minimizing the drop in machine vision analysis accuracy. In...
SourceID ieee
SourceType Publisher
StartPage 3411
SubjectTerms Codecs
finetuning
GANs
Image coding
Image coding for machines
Machine vision
Measurement
Surveillance
Training
VCM
Visualization
Title Bridging the Gap Between Image Coding for Machines and Humans
URI https://ieeexplore.ieee.org/document/9897916
WOSCitedRecordID wos001058109503101&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/eLvHCXMwlV1LawIxEB5UeuipDy19k0OPjWazZrO5VmrroeKhBW-SJ3joKq7293cSF0uhl95CYAh8IZlvkplvAB6896UKuaQB-S4d5p5TzbijyiA95lZ4E3RqNiGn03I-V7MWPB5qYdA4JZ_5fhymv3y3srv4VDZQpZJIZ9rQlrLY12odgquoG9NUAGdMDSajyWxYIJnGEJDzfmP5q4VK8iDjk_-tfQq9n1I8Mjs4mTNo-eocThruSJqTWXchqhO52HCIIKMjL3pNnvYZWGTyiVcGGa2iPUGKSt5S_qSvia4cSY_4dQ8-xs_vo1fatEagS87yLfWm9Fog-khndGlkph33ViLoggUpoiY718guHOeO2YJZg65fGBNyZWwWRH4BnWpV-UsguRZB66gTGGKwxbRQXBh0_E5n3Bl2Bd2IxmK9V79YNEBc_z19A8cR8Hi7Z8UtdLabnb-DI_u1Xdab-7Rl38Z7lwM
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1NTwIxEJ0gmugJFYzf9uDRQrdL2e1VIrIRCAdMuJF22yYeXAgL_n6nS4Mx8eJts8lkk9ds50078x7Ao7U2lS5OqEO-S7ux5VQxbqjUSI95Lqx2qjKbSCaTdD6X0xo87WdhMLhqPrNt_1jd5ZtlvvVHZR2ZygTpzAEceuesMK21L6-8ckyYAY6Y7GT9bNrtIZ3GIpDzdoj9ZaJS5ZBB439fP4XWzzAeme7TzBnUbHEOjcAeSfg3yyZ4fSLjLYcIcjryqlbkedeDRbJP3DRIf-njCZJUMq46KG1JVGFIdYxftuB98DLrD2kwR6AfnMUbanVqlUD8kdCoVCeRMtzmCcIumEuEV2XnCvmF4dywvMdyjclfaO1iqfPIifgC6sWysJdAYiWcUl4p0PlyiykhudCY-o2KuNHsCpoejcVqp3-xCEBc__36AY6Hs_FoMcombzdw4sH3e33Uu4X6Zr21d3CUf20-yvV9tXzf18-aTA
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%3Ajournal&rft.genre=proceeding&rft.title=Proceedings+-+International+Conference+on+Image+Processing&rft.atitle=Bridging+the+Gap+Between+Image+Coding+for+Machines+and+Humans&rft.au=Le%2C+Nam&rft.au=Zhang%2C+Honglei&rft.au=Cricri%2C+Francesco&rft.au=Youvalari%2C+Ramin+G.&rft.date=2022-10-16&rft.pub=IEEE&rft.eissn=2381-8549&rft.spage=3411&rft.epage=3415&rft_id=info:doi/10.1109%2FICIP46576.2022.9897916&rft.externalDocID=9897916