Channel-Wise Autoregressive Entropy Models for Learned Image Compression

In learning-based approaches to image compression, codecs are developed by optimizing a computational model to minimize a rate-distortion objective. Currently, the most effective learned image codecs take the form of an entropy-constrained autoencoder with an entropy model that uses both forward and...

Full description

Saved in:
Bibliographic Details
Published in:Proceedings - International Conference on Image Processing pp. 3339 - 3343
Main Authors: Minnen, David, Singh, Saurabh
Format: Conference Proceeding
Language:English
Published: IEEE 01.10.2020
Subjects:
Adaptation models
Adaptive Entropy Modeling
Bit rate
Codecs
Entropy
Image coding
Image Compression
Neural Networks
Predictive models
Training
ISSN:2381-8549
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Abstract In learning-based approaches to image compression, codecs are developed by optimizing a computational model to minimize a rate-distortion objective. Currently, the most effective learned image codecs take the form of an entropy-constrained autoencoder with an entropy model that uses both forward and backward adaptation. Forward adaptation makes use of side information and can be efficiently integrated into a deep neural network. In contrast, backward adaptation typically makes predictions based on the causal context of each symbol, which requires serial processing that prevents efficient GPU / TPU utilization. We introduce two enhancements, channel-conditioning and latent residual prediction, that lead to network architectures with better rate-distortion performance than existing context-adaptive models while minimizing serial processing. Empirically, we see an average rate savings of 6.7% on the Kodak image set and 11.4% on the Tecnick image set compared to a context-adaptive baseline model. At low bit rates, where the improvements are most effective, our model saves up to 18% over the baseline and outperforms hand-engineered codecs like BPG by up to 25%.
AbstractList In learning-based approaches to image compression, codecs are developed by optimizing a computational model to minimize a rate-distortion objective. Currently, the most effective learned image codecs take the form of an entropy-constrained autoencoder with an entropy model that uses both forward and backward adaptation. Forward adaptation makes use of side information and can be efficiently integrated into a deep neural network. In contrast, backward adaptation typically makes predictions based on the causal context of each symbol, which requires serial processing that prevents efficient GPU / TPU utilization. We introduce two enhancements, channel-conditioning and latent residual prediction, that lead to network architectures with better rate-distortion performance than existing context-adaptive models while minimizing serial processing. Empirically, we see an average rate savings of 6.7% on the Kodak image set and 11.4% on the Tecnick image set compared to a context-adaptive baseline model. At low bit rates, where the improvements are most effective, our model saves up to 18% over the baseline and outperforms hand-engineered codecs like BPG by up to 25%.
Author Minnen, David
Singh, Saurabh
Author_xml – sequence: 1
  givenname: David
  surname: Minnen
  fullname: Minnen, David
  organization: Google Research,Mountain View,CA,USA,94043
– sequence: 2
  givenname: Saurabh
  surname: Singh
  fullname: Singh, Saurabh
  organization: Google Research,Mountain View,CA,USA,94043
BookMark eNotj9FKwzAUQKMouM19gSD5gdbcm7ZJHkeZW2GiD4qPI11uZ6VNRjKF_b2iezovhwNnyq588MTYPYgcQJiHpm5eCqGUzlGgyA0YYWR5weZGaVCooZKmrC7ZBKWGTJeFuWHTlD7Frw0SJmxdf1jvacje-0R88XUMkfaRUuq_iS_9MYbDiT8FR0PiXYh8QzZ6crwZ7Z54HcbDnxz8Lbvu7JBofuaMvT0uX-t1tnleNfVik_Uo5DHDXWkAnXakWqM7g85Z4dq2cmQLQYhOYAdAqgCwViPuCpKqVa51nWutkTN299_tiWh7iP1o42l7Hpc_A3FQhA
ContentType Conference Proceeding
DBID 6IE
6IH
CBEJK
RIE
RIO
DOI 10.1109/ICIP40778.2020.9190935
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 9781728163956
1728163951
EISSN 2381-8549
EndPage 3343
ExternalDocumentID 9190935
Genre orig-research
GroupedDBID 29O
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-2c5912d8de7b98f92dda0dbb6dea40e22d02f11e7411aa822c4e37b7dbdfdba93
IEDL.DBID RIE
ISICitedReferencesCount 301
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000646178503090&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 02:34:00 EDT 2025
IsPeerReviewed false
IsScholarly true
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-i203t-2c5912d8de7b98f92dda0dbb6dea40e22d02f11e7411aa822c4e37b7dbdfdba93
PageCount 5
ParticipantIDs ieee_primary_9190935
PublicationCentury 2000
PublicationDate 2020-Oct.
PublicationDateYYYYMMDD 2020-10-01
PublicationDate_xml – month: 10
  year: 2020
  text: 2020-Oct.
PublicationDecade 2020
PublicationTitle Proceedings - International Conference on Image Processing
PublicationTitleAbbrev ICIP
PublicationYear 2020
Publisher IEEE
Publisher_xml – name: IEEE
SSID ssj0020131
Score 2.6346035
Snippet In learning-based approaches to image compression, codecs are developed by optimizing a computational model to minimize a rate-distortion objective. Currently,...
SourceID ieee
SourceType Publisher
StartPage 3339
SubjectTerms Adaptation models
Adaptive Entropy Modeling
Bit rate
Codecs
Entropy
Image coding
Image Compression
Neural Networks
Predictive models
Training
Title Channel-Wise Autoregressive Entropy Models for Learned Image Compression
URI https://ieeexplore.ieee.org/document/9190935
WOSCitedRecordID wos000646178503090&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/eLvHCXMwlV3Pa8IwFH6o7LCT23TsNznsuGqbNk1yHKIoDPGwMW-SNC8guCq2DvbfL2mLY7DLbiGQBF7Je99Lv-89gEfBhHK-Nw60zFyCQpkNhPTtXjQ1VjpImtpKKPzC53OxXMpFC56OWhhErMhnOPDD6l--2WYH_1Q2lC56yZi1oc15Wmu1jsmVrxvTKICjUA5no9nC5Srcs7doOGhW_mqhUkWQSfd_Z59B_0eKRxbHIHMOLcwvoNtgR9LczKIHUy8TyHETvK8LJM--MgFWqbTzZmTs6ei7L-Ibn20K4nAqqeqquh1mH86hEO8VakJs3oe3yfh1NA2aLgnBmoZxGdCMyYgaYZBrKaykxqjQaJ0aVEmIlJqQ2ihCBx0ipRweyBKMueZGG2u0kvEldPJtjldA0lRFNk5UGIskYVZoxlVE04wpq5GhuYaeN8xqVxfCWDU2ufl7-hZOve1r5tsddMr9Ae_hJPss18X-ofp63ya0nQc
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV3Pa8IwFH44N9hObtOx38thx1XbtGmT4xBFmRMPjnmTpHkBwVWxOth_v6QtjsEuu4VAEkjIe99Lvu89gEfOuLS2N_SUSG2AQpnxuHDlXhTVRlhIGptCKDxKxmM-m4lJDZ72WhhELMhn2HbN4i9fr9KdeyrrCOu9RMgO4JBZP8pLtdY-vHKZYyoNcOCLzrA7nNhoJXH8Leq3q7G_iqgUPqTf-N_qp9D6EeORyd7NnEENs3NoVOiRVHczb8LACQUyXHrvixzJs8tNgEUwbe0Z6TlC-vqLuNJny5xYpEqKzKp2huGHNSnE2YWSEpu14K3fm3YHXlUnwVtQP9x6NGUioJprTJTgRlCtpa-VijXKyEdKtU9NEKAFD4GUFhGkEYaJSrTSRispwguoZ6sML4HEsQxMGEk_5FHEDFcskQGNUyaNQob6CppuY-brMhXGvNqT67-7H-B4MH0dzUfD8csNnLhzKHlwt1DfbnZ4B0fp53aRb-6Lk_wGmbmgXA
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+-+International+Conference+on+Image+Processing&rft.atitle=Channel-Wise+Autoregressive+Entropy+Models+for+Learned+Image+Compression&rft.au=Minnen%2C+David&rft.au=Singh%2C+Saurabh&rft.date=2020-10-01&rft.pub=IEEE&rft.eissn=2381-8549&rft.spage=3339&rft.epage=3343&rft_id=info:doi/10.1109%2FICIP40778.2020.9190935&rft.externalDocID=9190935