Learning Scalable $\ell_\infty$-constrained Near-lossless Image Compression via Joint Lossy Image and Residual Compression Article Swipe

PDF

Yuanchao Bai , Xianming Liu , Wangmeng Zuo , Yaowei Wang , Xiangyang Ji ·

YOU? · · 2021 · Open Access · · DOI: https://doi.org/10.48550/arxiv.2103.17015

We propose a novel joint lossy image and residual compression framework for learning $\ell_\infty$-constrained near-lossless image compression. Specifically, we obtain a lossy reconstruction of the raw image through lossy image compression and uniformly quantize the corresponding residual to satisfy a given tight $\ell_\infty$ error bound. Suppose that the error bound is zero, i.e., lossless image compression, we formulate the joint optimization problem of compressing both the lossy image and the original residual in terms of variational auto-encoders and solve it with end-to-end training. To achieve scalable compression with the error bound larger than zero, we derive the probability model of the quantized residual by quantizing the learned probability model of the original residual, instead of training multiple networks. We further correct the bias of the derived probability model caused by the context mismatch between training and inference. Finally, the quantized residual is encoded according to the bias-corrected probability model and is concatenated with the bitstream of the compressed lossy image. Experimental results demonstrate that our near-lossless codec achieves the state-of-the-art performance for lossless and near-lossless image compression, and achieves competitive PSNR while much smaller $\ell_\infty$ error compared with lossy image codecs at high bit rates.

Related Topics

Artificial Intelligence

Concepts

Lossy compression Lossless compression Image compression Codec Residual Algorithm Computer science Data compression Data compression ratio Compression artifact Mathematics Artificial intelligence Image (mathematics) Image processing Telecommunications

Metadata

Type: preprint
Language: en
Landing Page: http://arxiv.org/abs/2103.17015
PDF: https://arxiv.org/pdf/2103.17015
OA Status: green
Cited By: 1
References: 42
Related Works: 10
OpenAlex ID: https://openalex.org/W3142718519

All OpenAlex metadata

Raw OpenAlex JSON

OpenAlex ID: https://openalex.org/W3142718519

Canonical identifier for this work in OpenAlex
DOI: https://doi.org/10.48550/arxiv.2103.17015

Digital Object Identifier
Title: Learning Scalable $\ell_\infty$-constrained Near-lossless Image Compression via Joint Lossy Image and Residual Compression

Work title
Type: preprint

OpenAlex work type
Language: en

Primary language
Publication year: 2021

Year of publication
Publication date: 2021-03-31

Full publication date if available
Authors: Yuanchao Bai, Xianming Liu, Wangmeng Zuo, Yaowei Wang, Xiangyang Ji

List of authors in order
Landing page: https://arxiv.org/abs/2103.17015

Publisher landing page
PDF URL: https://arxiv.org/pdf/2103.17015

Direct link to full text PDF
Open access: Yes

Whether a free full text is available
OA status: green

Open access status per OpenAlex
OA URL: https://arxiv.org/pdf/2103.17015

Direct OA link when available
Concepts: Lossy compression, Lossless compression, Image compression, Codec, Residual, Algorithm, Computer science, Data compression, Data compression ratio, Compression artifact, Mathematics, Artificial intelligence, Image (mathematics), Image processing, Telecommunications

Top concepts (fields/topics) attached by OpenAlex
Cited by: 1

Total citation count in OpenAlex
Citations by year (recent): 2021: 1

Per-year citation counts (last 5 years)
References (count): 42

Number of works referenced by this work
Related works (count): 10

Other works algorithmically related by OpenAlex

Full payload

id	https://openalex.org/W3142718519
doi	https://doi.org/10.48550/arxiv.2103.17015
ids.doi	https://doi.org/10.48550/arxiv.2103.17015
ids.mag	3142718519
ids.openalex	https://openalex.org/W3142718519
fwci
type	preprint
title	Learning Scalable $\ell_\infty$-constrained Near-lossless Image Compression via Joint Lossy Image and Residual Compression
biblio.issue
biblio.volume
biblio.last_page
biblio.first_page
topics[0].id	https://openalex.org/T11105
topics[0].field.id	https://openalex.org/fields/17
topics[0].field.display_name	Computer Science
topics[0].score	0.9998999834060669
topics[0].domain.id	https://openalex.org/domains/3
topics[0].domain.display_name	Physical Sciences
topics[0].subfield.id	https://openalex.org/subfields/1707
topics[0].subfield.display_name	Computer Vision and Pattern Recognition
topics[0].display_name	Advanced Image Processing Techniques
topics[1].id	https://openalex.org/T10901
topics[1].field.id	https://openalex.org/fields/17
topics[1].field.display_name	Computer Science
topics[1].score	0.9998999834060669
topics[1].domain.id	https://openalex.org/domains/3
topics[1].domain.display_name	Physical Sciences
topics[1].subfield.id	https://openalex.org/subfields/1707
topics[1].subfield.display_name	Computer Vision and Pattern Recognition
topics[1].display_name	Advanced Data Compression Techniques
topics[2].id	https://openalex.org/T10688
topics[2].field.id	https://openalex.org/fields/17
topics[2].field.display_name	Computer Science
topics[2].score	0.9998000264167786
topics[2].domain.id	https://openalex.org/domains/3
topics[2].domain.display_name	Physical Sciences
topics[2].subfield.id	https://openalex.org/subfields/1707
topics[2].subfield.display_name	Computer Vision and Pattern Recognition
topics[2].display_name	Image and Signal Denoising Methods
is_xpac	False
apc_list
apc_paid
concepts[0].id	https://openalex.org/C165021410
concepts[0].level	2
concepts[0].score	0.9327009916305542
concepts[0].wikidata	https://www.wikidata.org/wiki/Q55564
concepts[0].display_name	Lossy compression
concepts[1].id	https://openalex.org/C81081738
concepts[1].level	3
concepts[1].score	0.8190199732780457
concepts[1].wikidata	https://www.wikidata.org/wiki/Q55542
concepts[1].display_name	Lossless compression
concepts[2].id	https://openalex.org/C13481523
concepts[2].level	4
concepts[2].score	0.7404832243919373
concepts[2].wikidata	https://www.wikidata.org/wiki/Q412438
concepts[2].display_name	Image compression
concepts[3].id	https://openalex.org/C161765866
concepts[3].level	2
concepts[3].score	0.5876967906951904
concepts[3].wikidata	https://www.wikidata.org/wiki/Q184748
concepts[3].display_name	Codec
concepts[4].id	https://openalex.org/C155512373
concepts[4].level	2
concepts[4].score	0.5809244513511658
concepts[4].wikidata	https://www.wikidata.org/wiki/Q287450
concepts[4].display_name	Residual
concepts[5].id	https://openalex.org/C11413529
concepts[5].level	1
concepts[5].score	0.5414711236953735
concepts[5].wikidata	https://www.wikidata.org/wiki/Q8366
concepts[5].display_name	Algorithm
concepts[6].id	https://openalex.org/C41008148
concepts[6].level	0
concepts[6].score	0.496401846408844
concepts[6].wikidata	https://www.wikidata.org/wiki/Q21198
concepts[6].display_name	Computer science
concepts[7].id	https://openalex.org/C78548338
concepts[7].level	2
concepts[7].score	0.48638680577278137
concepts[7].wikidata	https://www.wikidata.org/wiki/Q2493
concepts[7].display_name	Data compression
concepts[8].id	https://openalex.org/C94835093
concepts[8].level	5
concepts[8].score	0.4830799400806427
concepts[8].wikidata	https://www.wikidata.org/wiki/Q3113333
concepts[8].display_name	Data compression ratio
concepts[9].id	https://openalex.org/C57654395
concepts[9].level	5
concepts[9].score	0.41712307929992676
concepts[9].wikidata	https://www.wikidata.org/wiki/Q1097775
concepts[9].display_name	Compression artifact
concepts[10].id	https://openalex.org/C33923547
concepts[10].level	0
concepts[10].score	0.39829832315444946
concepts[10].wikidata	https://www.wikidata.org/wiki/Q395
concepts[10].display_name	Mathematics
concepts[11].id	https://openalex.org/C154945302
concepts[11].level	1
concepts[11].score	0.3791336715221405
concepts[11].wikidata	https://www.wikidata.org/wiki/Q11660
concepts[11].display_name	Artificial intelligence
concepts[12].id	https://openalex.org/C115961682
concepts[12].level	2
concepts[12].score	0.3205854892730713
concepts[12].wikidata	https://www.wikidata.org/wiki/Q860623
concepts[12].display_name	Image (mathematics)
concepts[13].id	https://openalex.org/C9417928
concepts[13].level	3
concepts[13].score	0.18549895286560059
concepts[13].wikidata	https://www.wikidata.org/wiki/Q1070689
concepts[13].display_name	Image processing
concepts[14].id	https://openalex.org/C76155785
concepts[14].level	1
concepts[14].score	0.05843457579612732
concepts[14].wikidata	https://www.wikidata.org/wiki/Q418
concepts[14].display_name	Telecommunications
keywords[0].id	https://openalex.org/keywords/lossy-compression
keywords[0].score	0.9327009916305542
keywords[0].display_name	Lossy compression
keywords[1].id	https://openalex.org/keywords/lossless-compression
keywords[1].score	0.8190199732780457
keywords[1].display_name	Lossless compression
keywords[2].id	https://openalex.org/keywords/image-compression
keywords[2].score	0.7404832243919373
keywords[2].display_name	Image compression
keywords[3].id	https://openalex.org/keywords/codec
keywords[3].score	0.5876967906951904
keywords[3].display_name	Codec
keywords[4].id	https://openalex.org/keywords/residual
keywords[4].score	0.5809244513511658
keywords[4].display_name	Residual
keywords[5].id	https://openalex.org/keywords/algorithm
keywords[5].score	0.5414711236953735
keywords[5].display_name	Algorithm
keywords[6].id	https://openalex.org/keywords/computer-science
keywords[6].score	0.496401846408844
keywords[6].display_name	Computer science
keywords[7].id	https://openalex.org/keywords/data-compression
keywords[7].score	0.48638680577278137
keywords[7].display_name	Data compression
keywords[8].id	https://openalex.org/keywords/data-compression-ratio
keywords[8].score	0.4830799400806427
keywords[8].display_name	Data compression ratio
keywords[9].id	https://openalex.org/keywords/compression-artifact
keywords[9].score	0.41712307929992676
keywords[9].display_name	Compression artifact
keywords[10].id	https://openalex.org/keywords/mathematics
keywords[10].score	0.39829832315444946
keywords[10].display_name	Mathematics
keywords[11].id	https://openalex.org/keywords/artificial-intelligence
keywords[11].score	0.3791336715221405
keywords[11].display_name	Artificial intelligence
keywords[12].id	https://openalex.org/keywords/image
keywords[12].score	0.3205854892730713
keywords[12].display_name	Image (mathematics)
keywords[13].id	https://openalex.org/keywords/image-processing
keywords[13].score	0.18549895286560059
keywords[13].display_name	Image processing
keywords[14].id	https://openalex.org/keywords/telecommunications
keywords[14].score	0.05843457579612732
keywords[14].display_name	Telecommunications
language	en
locations[0].id	pmh:oai:arXiv.org:2103.17015
locations[0].is_oa	True
locations[0].source.id	https://openalex.org/S4306400194
locations[0].source.issn
locations[0].source.type	repository
locations[0].source.is_oa	True
locations[0].source.issn_l
locations[0].source.is_core	False
locations[0].source.is_in_doaj	False
locations[0].source.display_name	arXiv (Cornell University)
locations[0].source.host_organization	https://openalex.org/I205783295
locations[0].source.host_organization_name	Cornell University
locations[0].source.host_organization_lineage	https://openalex.org/I205783295
locations[0].license
locations[0].pdf_url	https://arxiv.org/pdf/2103.17015
locations[0].version	submittedVersion
locations[0].raw_type
locations[0].license_id
locations[0].is_accepted	False
locations[0].is_published	False
locations[0].raw_source_name
locations[0].landing_page_url	http://arxiv.org/abs/2103.17015
locations[1].id	doi:10.48550/arxiv.2103.17015
locations[1].is_oa	True
locations[1].source.id	https://openalex.org/S4306400194
locations[1].source.issn
locations[1].source.type	repository
locations[1].source.is_oa	True
locations[1].source.issn_l
locations[1].source.is_core	False
locations[1].source.is_in_doaj	False
locations[1].source.display_name	arXiv (Cornell University)
locations[1].source.host_organization	https://openalex.org/I205783295
locations[1].source.host_organization_name	Cornell University
locations[1].source.host_organization_lineage	https://openalex.org/I205783295
locations[1].license
locations[1].pdf_url
locations[1].version
locations[1].raw_type	article
locations[1].license_id
locations[1].is_accepted	False
locations[1].is_published
locations[1].raw_source_name
locations[1].landing_page_url	https://doi.org/10.48550/arxiv.2103.17015
indexed_in	arxiv, datacite
authorships[0].author.id	https://openalex.org/A5024093994
authorships[0].author.orcid	https://orcid.org/0000-0003-3449-6537
authorships[0].author.display_name	Yuanchao Bai
authorships[0].author_position	first
authorships[0].raw_author_name	Yuanchao Bai
authorships[0].is_corresponding	False
authorships[1].author.id	https://openalex.org/A5100654390
authorships[1].author.orcid	https://orcid.org/0000-0002-8857-1785
authorships[1].author.display_name	Xianming Liu
authorships[1].author_position	middle
authorships[1].raw_author_name	Xianming Liu
authorships[1].is_corresponding	False
authorships[2].author.id	https://openalex.org/A5100636655
authorships[2].author.orcid	https://orcid.org/0000-0002-3330-783X
authorships[2].author.display_name	Wangmeng Zuo
authorships[2].author_position	middle
authorships[2].raw_author_name	Wangmeng Zuo
authorships[2].is_corresponding	False
authorships[3].author.id	https://openalex.org/A5100631216
authorships[3].author.orcid	https://orcid.org/0000-0003-2197-9038
authorships[3].author.display_name	Yaowei Wang
authorships[3].author_position	middle
authorships[3].raw_author_name	Yaowei Wang
authorships[3].is_corresponding	False
authorships[4].author.id	https://openalex.org/A5024401174
authorships[4].author.orcid	https://orcid.org/0000-0002-7333-9975
authorships[4].author.display_name	Xiangyang Ji
authorships[4].author_position	last
authorships[4].raw_author_name	Xiangyang Ji
authorships[4].is_corresponding	False
has_content.pdf	False
has_content.grobid_xml	False
is_paratext	False
open_access.is_oa	True
open_access.oa_url	https://arxiv.org/pdf/2103.17015
open_access.oa_status	green
open_access.any_repository_has_fulltext	False
created_date	2021-04-13T00:00:00
display_name	Learning Scalable $\ell_\infty$-constrained Near-lossless Image Compression via Joint Lossy Image and Residual Compression
has_fulltext	False
is_retracted	False
updated_date	2025-11-06T06:51:31.235846
primary_topic.id	https://openalex.org/T11105
primary_topic.field.id	https://openalex.org/fields/17
primary_topic.field.display_name	Computer Science
primary_topic.score	0.9998999834060669
primary_topic.domain.id	https://openalex.org/domains/3
primary_topic.domain.display_name	Physical Sciences
primary_topic.subfield.id	https://openalex.org/subfields/1707
primary_topic.subfield.display_name	Computer Vision and Pattern Recognition
primary_topic.display_name	Advanced Image Processing Techniques
related_works	https://openalex.org/W4287393224, https://openalex.org/W3118996461, https://openalex.org/W2541319825, https://openalex.org/W2071387875, https://openalex.org/W1680283075, https://openalex.org/W3080614128, https://openalex.org/W4239869440, https://openalex.org/W2357670775, https://openalex.org/W2380116549, https://openalex.org/W3202870363
cited_by_count	1
counts_by_year[0].year	2021
counts_by_year[0].cited_by_count	1
locations_count	2
best_oa_location.id	pmh:oai:arXiv.org:2103.17015
best_oa_location.is_oa	True
best_oa_location.source.id	https://openalex.org/S4306400194
best_oa_location.source.issn
best_oa_location.source.type	repository
best_oa_location.source.is_oa	True
best_oa_location.source.issn_l
best_oa_location.source.is_core	False
best_oa_location.source.is_in_doaj	False
best_oa_location.source.display_name	arXiv (Cornell University)
best_oa_location.source.host_organization	https://openalex.org/I205783295
best_oa_location.source.host_organization_name	Cornell University
best_oa_location.source.host_organization_lineage	https://openalex.org/I205783295
best_oa_location.license
best_oa_location.pdf_url	https://arxiv.org/pdf/2103.17015
best_oa_location.version	submittedVersion
best_oa_location.raw_type
best_oa_location.license_id
best_oa_location.is_accepted	False
best_oa_location.is_published	False
best_oa_location.raw_source_name
best_oa_location.landing_page_url	http://arxiv.org/abs/2103.17015
primary_location.id	pmh:oai:arXiv.org:2103.17015
primary_location.is_oa	True
primary_location.source.id	https://openalex.org/S4306400194
primary_location.source.issn
primary_location.source.type	repository
primary_location.source.is_oa	True
primary_location.source.issn_l
primary_location.source.is_core	False
primary_location.source.is_in_doaj	False
primary_location.source.display_name	arXiv (Cornell University)
primary_location.source.host_organization	https://openalex.org/I205783295
primary_location.source.host_organization_name	Cornell University
primary_location.source.host_organization_lineage	https://openalex.org/I205783295
primary_location.license
primary_location.pdf_url	https://arxiv.org/pdf/2103.17015
primary_location.version	submittedVersion
primary_location.raw_type
primary_location.license_id
primary_location.is_accepted	False
primary_location.is_published	False
primary_location.raw_source_name
primary_location.landing_page_url	http://arxiv.org/abs/2103.17015
publication_date	2021-03-31
publication_year	2021
referenced_works	https://openalex.org/W2153638435, https://openalex.org/W2963948038, https://openalex.org/W2099563019, https://openalex.org/W3014578089, https://openalex.org/W2096445898, https://openalex.org/W1518500973, https://openalex.org/W2166816543, https://openalex.org/W4226151124, https://openalex.org/W2129652681, https://openalex.org/W2099778990, https://openalex.org/W2970774025, https://openalex.org/W2962891349, https://openalex.org/W2965631471, https://openalex.org/W2893920456, https://openalex.org/W2963449488, https://openalex.org/W135467536, https://openalex.org/W2962790638, https://openalex.org/W1580389772, https://openalex.org/W2140196014, https://openalex.org/W2133665775, https://openalex.org/W1959608418, https://openalex.org/W2518435514, https://openalex.org/W1995875735, https://openalex.org/W2741137940, https://openalex.org/W2964098744, https://openalex.org/W2892278106, https://openalex.org/W2964122153, https://openalex.org/W2982853315, https://openalex.org/W2011512259, https://openalex.org/W2963636093, https://openalex.org/W3103781353, https://openalex.org/W3103543904, https://openalex.org/W2964121744, https://openalex.org/W2267126114, https://openalex.org/W2110707318, https://openalex.org/W2963149687, https://openalex.org/W3034743539, https://openalex.org/W2954655735, https://openalex.org/W3035294152, https://openalex.org/W2593493485, https://openalex.org/W3034469748, https://openalex.org/W2604392022
referenced_works_count	42
abstract_inverted_index.a	2, 20, 39
abstract_inverted_index.To	83
abstract_inverted_index.We	0, 118
abstract_inverted_index.at	191
abstract_inverted_index.by	103, 129
abstract_inverted_index.in	72
abstract_inverted_index.is	50, 141, 150
abstract_inverted_index.it	79
abstract_inverted_index.of	23, 62, 74, 99, 109, 114, 123, 155
abstract_inverted_index.to	37, 144
abstract_inverted_index.we	18, 56, 94
abstract_inverted_index.and	7, 31, 68, 77, 135, 149, 173, 177
abstract_inverted_index.bit	193
abstract_inverted_index.for	11, 171
abstract_inverted_index.our	164
abstract_inverted_index.raw	25
abstract_inverted_index.the	24, 34, 47, 58, 65, 69, 88, 96, 100, 105, 110, 121, 124, 130, 138, 145, 153, 156, 168
abstract_inverted_index.PSNR	180
abstract_inverted_index.bias	122
abstract_inverted_index.both	64
abstract_inverted_index.high	192
abstract_inverted_index.much	182
abstract_inverted_index.than	92
abstract_inverted_index.that	46, 163
abstract_inverted_index.with	80, 87, 152, 187
abstract_inverted_index.bound	49, 90
abstract_inverted_index.codec	166
abstract_inverted_index.error	43, 48, 89, 185
abstract_inverted_index.given	40
abstract_inverted_index.i.e.,	52
abstract_inverted_index.image	6, 15, 26, 29, 54, 67, 175, 189
abstract_inverted_index.joint	4, 59
abstract_inverted_index.lossy	5, 21, 28, 66, 158, 188
abstract_inverted_index.model	98, 108, 127, 148
abstract_inverted_index.novel	3
abstract_inverted_index.solve	78
abstract_inverted_index.terms	73
abstract_inverted_index.tight	41
abstract_inverted_index.while	181
abstract_inverted_index.zero,	51, 93
abstract_inverted_index.bound.	44
abstract_inverted_index.caused	128
abstract_inverted_index.codecs	190
abstract_inverted_index.derive	95
abstract_inverted_index.image.	159
abstract_inverted_index.larger	91
abstract_inverted_index.obtain	19
abstract_inverted_index.rates.	194
abstract_inverted_index.Suppose	45
abstract_inverted_index.achieve	84
abstract_inverted_index.between	133
abstract_inverted_index.context	131
abstract_inverted_index.correct	120
abstract_inverted_index.derived	125
abstract_inverted_index.encoded	142
abstract_inverted_index.further	119
abstract_inverted_index.instead	113
abstract_inverted_index.learned	106
abstract_inverted_index.problem	61
abstract_inverted_index.propose	1
abstract_inverted_index.results	161
abstract_inverted_index.satisfy	38
abstract_inverted_index.smaller	183
abstract_inverted_index.through	27
abstract_inverted_index.Finally,	137
abstract_inverted_index.achieves	167, 178
abstract_inverted_index.compared	186
abstract_inverted_index.learning	12
abstract_inverted_index.lossless	53, 172
abstract_inverted_index.mismatch	132
abstract_inverted_index.multiple	116
abstract_inverted_index.original	70, 111
abstract_inverted_index.quantize	33
abstract_inverted_index.residual	8, 36, 71, 102, 140
abstract_inverted_index.scalable	85
abstract_inverted_index.training	115, 134
abstract_inverted_index.according	143
abstract_inverted_index.bitstream	154
abstract_inverted_index.formulate	57
abstract_inverted_index.framework	10
abstract_inverted_index.networks.	117
abstract_inverted_index.quantized	101, 139
abstract_inverted_index.residual,	112
abstract_inverted_index.training.	82
abstract_inverted_index.uniformly	32
abstract_inverted_index.compressed	157
abstract_inverted_index.end-to-end	81
abstract_inverted_index.inference.	136
abstract_inverted_index.quantizing	104
abstract_inverted_index.competitive	179
abstract_inverted_index.compressing	63
abstract_inverted_index.compression	9, 30, 86
abstract_inverted_index.demonstrate	162
abstract_inverted_index.performance	170
abstract_inverted_index.probability	97, 107, 126, 147
abstract_inverted_index.variational	75
abstract_inverted_index.Experimental	160
abstract_inverted_index.compression,	55, 176
abstract_inverted_index.compression.	16
abstract_inverted_index.concatenated	151
abstract_inverted_index.optimization	60
abstract_inverted_index.$\ell_\infty$	42, 184
abstract_inverted_index.Specifically,	17
abstract_inverted_index.auto-encoders	76
abstract_inverted_index.corresponding	35
abstract_inverted_index.near-lossless	14, 165, 174
abstract_inverted_index.bias-corrected	146
abstract_inverted_index.reconstruction	22
abstract_inverted_index.state-of-the-art	169
abstract_inverted_index.$\ell_\infty$-constrained	13
cited_by_percentile_year
countries_distinct_count	0
institutions_distinct_count	5
citation_normalized_percentile