Delta-ICM: Entropy Modeling with Delta Function for Learned Image Compression

Image Coding for Machines (ICM) is becoming more important as research in computer vision progresses. ICM is a vital research field that pursues the use of images for image recognition models, facilitating efficient image transmission and storage. Demand and required performance for recognition mode...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:Proceedings of IEEE International Symposium on Consumer Electronics s. 1 - 6
Hlavní autori: Shindo, Takahiro, Watanabe, Taiju, Tatsumi, Yui, Watanabe, Hiroshi
Médium: Konferenčný príspevok..
Jazyk:English
Vydavateľské údaje: IEEE 11.01.2025
Predmet:
ISSN:2158-4001
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Abstract Image Coding for Machines (ICM) is becoming more important as research in computer vision progresses. ICM is a vital research field that pursues the use of images for image recognition models, facilitating efficient image transmission and storage. Demand and required performance for recognition models are rapidly growing within consumers. To meet these needs, exchanging image data between consumer devices and cloud AI using ICM technology could be one possible solution. In ICM, various image compression methods have adopted Learned Image Compression (LIC). LIC includes an entropy model for estimating the bitrate of latent features, and the design of this model significantly affects its performance. Typically, LIC methods assume that the distribution of latent features follows a normal distribution. This assumption is effective for compressing images intended for human vision. However, employing an entropy model based on normal distribution is inefficient in ICM due to the limitation of image parts that require precise decoding. To address this, we propose Delta-ICM, which uses a probability distribution based on a delta function. Assuming the delta distribution as a distribution of latent features reduces the entropy of image portions unnecessary for machines. We compress the remaining portions using an entropy model based on normal distribution, similar to existing methods. Delta-ICM selects between the entropy model based on the delta distribution and the one based on the normal distribution for each latent feature. Our method outperforms existing ICM methods in image compression performance aimed at machines.
AbstractList Image Coding for Machines (ICM) is becoming more important as research in computer vision progresses. ICM is a vital research field that pursues the use of images for image recognition models, facilitating efficient image transmission and storage. Demand and required performance for recognition models are rapidly growing within consumers. To meet these needs, exchanging image data between consumer devices and cloud AI using ICM technology could be one possible solution. In ICM, various image compression methods have adopted Learned Image Compression (LIC). LIC includes an entropy model for estimating the bitrate of latent features, and the design of this model significantly affects its performance. Typically, LIC methods assume that the distribution of latent features follows a normal distribution. This assumption is effective for compressing images intended for human vision. However, employing an entropy model based on normal distribution is inefficient in ICM due to the limitation of image parts that require precise decoding. To address this, we propose Delta-ICM, which uses a probability distribution based on a delta function. Assuming the delta distribution as a distribution of latent features reduces the entropy of image portions unnecessary for machines. We compress the remaining portions using an entropy model based on normal distribution, similar to existing methods. Delta-ICM selects between the entropy model based on the delta distribution and the one based on the normal distribution for each latent feature. Our method outperforms existing ICM methods in image compression performance aimed at machines.
Author Shindo, Takahiro
Watanabe, Taiju
Tatsumi, Yui
Watanabe, Hiroshi
Author_xml – sequence: 1
  givenname: Takahiro
  surname: Shindo
  fullname: Shindo, Takahiro
  email: taka_s0265@ruri.waseda.jp
  organization: Waseda University,Tokyo,Japan
– sequence: 2
  givenname: Taiju
  surname: Watanabe
  fullname: Watanabe, Taiju
  email: lvpurin@fuji.waseda.jp
  organization: Waseda University,Tokyo,Japan
– sequence: 3
  givenname: Yui
  surname: Tatsumi
  fullname: Tatsumi, Yui
  email: yui.t@fuji.waseda.jp
  organization: Waseda University,Tokyo,Japan
– sequence: 4
  givenname: Hiroshi
  surname: Watanabe
  fullname: Watanabe, Hiroshi
  email: hiroshi.watanabe@waseda.jp
  organization: Waseda University,Tokyo,Japan
BookMark eNo1j9tKxDAYhKMouLvuGwjmBVpzbBLvJHa10OLN3i9p82etbNPSVmTf3uLhaob5hoFZo6vYR0DonpKUUmIeCmvzjGdCpYwwmS4RM1qwC7Q1ymjOqWSUZvISrRiVOhGE0Bu0nqaPxRgjzQpVz3CaXVLY6hHncR774Yyr3sOpjUf81c7v-KeAd5-xmds-4tCPuAQ3RvC46NwRsO27YYRpWugtug7uNMH2Tzdov8v39jUp314K-1QmreFz4pQSTXCqlhSCbwjomvoQhCGuNhK0U40QWnDgxntvKDimhIbgWAZKBMU36O53tgWAwzC2nRvPh__3_BtSdFF2
ContentType Conference Proceeding
DBID 6IE
6IH
CBEJK
RIE
RIO
DOI 10.1109/ICCE63647.2025.10929842
DatabaseName IEEE Electronic Library (IEL) Conference Proceedings
IEEE Proceedings Order Plan (POP) 1998-present by volume
IEEE Xplore All Conference Proceedings
IEEE Xplore Digital Library
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 9798331521165
EISSN 2158-4001
EndPage 6
ExternalDocumentID 10929842
Genre orig-research
GrantInformation_xml – fundername: National Institute of Information and Communications Technology (NICT), Japan
  grantid: JPJ012368C05101
  funderid: 10.13039/501100012389
GroupedDBID 6IE
6IF
6IH
6IL
6IN
AAWTH
ABLEC
ACGFS
ADZIZ
ALMA_UNASSIGNED_HOLDINGS
BEFXN
BFFAM
BGNUA
BKEBE
BPEOZ
CBEJK
CHZPO
IEGSK
IJVOP
M43
OCL
RIE
RIL
RIO
ID FETCH-LOGICAL-i93t-a774cfa7b51efdc0e8b1dff490ab95e8a7c44843e39ddd91ea2748efa26e74f73
IEDL.DBID RIE
IngestDate Wed Aug 27 01:41:05 EDT 2025
IsPeerReviewed true
IsScholarly true
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-i93t-a774cfa7b51efdc0e8b1dff490ab95e8a7c44843e39ddd91ea2748efa26e74f73
PageCount 6
ParticipantIDs ieee_primary_10929842
PublicationCentury 2000
PublicationDate 2025-Jan.-11
PublicationDateYYYYMMDD 2025-01-11
PublicationDate_xml – month: 01
  year: 2025
  text: 2025-Jan.-11
  day: 11
PublicationDecade 2020
PublicationTitle Proceedings of IEEE International Symposium on Consumer Electronics
PublicationTitleAbbrev ICCE
PublicationYear 2025
Publisher IEEE
Publisher_xml – name: IEEE
SSID ssj0019959
Score 2.2933636
Snippet Image Coding for Machines (ICM) is becoming more important as research in computer vision progresses. ICM is a vital research field that pursues the use of...
SourceID ieee
SourceType Publisher
StartPage 1
SubjectTerms Computational modeling
Delta Distribution
Entropy
Gaussian distribution
Image coding
Image Coding for Machines
Image recognition
Image texture
Learned Image Compression
Performance evaluation
Probability distribution
Smart devices
YOLO
Title Delta-ICM: Entropy Modeling with Delta Function for Learned Image Compression
URI https://ieeexplore.ieee.org/document/10929842
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LSwMxEA4qHvTiq-KbHLymNs1uHl7XFgu29NBDbyXZTKCg21K3gv_eSfpQDx68hZAhMCEzmcl83xByL6xEyytKJq03LDLKMedlYEJ454M1JneJZ_ZFDQZ6PDbDNVg9YWEAIBWfQTMO01--n5XLmCrDG47OXGdocXeVkiuw1vbLIBJnrQu4cN1Dryg6MpKjYwzYzpsb0V9NVJIP6R79c_dj0vhG49Hh1s-ckB2oTsnhDyLBM9J_gtfasl7Rf6SdWHs-_6Sxy1nEmtOYaqVpAe2iF4snQfGpShO1Knjae0ObQqNhWNXEVg0y6nZGxTNbN0pgUyNqZvEJVwarXM4h-LIF2nEfQmZa1pkctFUlBmGZAGG894aDxVBUQ7BtCSoLSpyTvWpWwQWhmc3BaVQgDxplrBEe44lEFgpc6vYlaUTFTOYrKozJRidXf8xfk4Oo_piz4PyG7NWLJdyS_fKjnr4v7tIBfgGsR5yu
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LTwIxEG4MmqgXXxjf9uC1yNJ9tF4RwkYgHDhwI93tNCHRheBi4r93piyoBw_emqZNk5l0pjOd7xvGHqSJ0fLKXMTGakGMciKzsRNS2sw6o3WUeZ7ZfjIcqslEjyqwusfCAIAvPoMGDf1fvp3nK0qV4Q1HZ65CtLi71DqrgmttPw2IOqsq4cKVj2m73YmJHh2jwFbU2Gz-1UbFe5Hu0T_PP2b1bzweH209zQnbgeKUHf6gEjxjg2d4LY1I24Mn3qHq88Unpz5nhDbnlGzlfgHvoh8jXXB8rHJPrgqWp29oVTiZhnVVbFFn425n3O6JqlWCmGlZCoOPuNyZJIsCcDZvgsoC61yomybTESiT5BiGhRKkttbqAAwGowqcacWQhC6R56xWzAu4YDw0EWQKBRg4hXuMlhYjCk8XCkGsWpesToKZLtZkGNONTK7-mL9n-73xoD_tp8OXa3ZAqqAMRhDcsFq5XMEt28s_ytn78s4r8wsICp_3
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+IEEE+International+Symposium+on+Consumer+Electronics&rft.atitle=Delta-ICM%3A+Entropy+Modeling+with+Delta+Function+for+Learned+Image+Compression&rft.au=Shindo%2C+Takahiro&rft.au=Watanabe%2C+Taiju&rft.au=Tatsumi%2C+Yui&rft.au=Watanabe%2C+Hiroshi&rft.date=2025-01-11&rft.pub=IEEE&rft.eissn=2158-4001&rft.spage=1&rft.epage=6&rft_id=info:doi/10.1109%2FICCE63647.2025.10929842&rft.externalDocID=10929842