Lossless Compression of Point Cloud Sequences Using Sequence Optimized CNN Models

In this paper we propose a new paradigm for encoding the geometry of dense point cloud sequences, where a convolutional neural network (CNN), which estimates the encoding distributions, is optimized on several frames of the sequence to be compressed. We adopt lightweight CNN structures, we perform t...

Celý popis

Uložené v:

Podrobná bibliografia
Vydané v:	IEEE access Ročník 10; s. 83678 - 83691
Hlavní autori:	Kaya, Emre C., Tabus, Ioan
Médium:	Journal Article
Jazyk:	English
Vydavateľské údaje:	Piscataway IEEE 2022 The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Predmet:	Arithmetic coding Artificial neural networks Context modeling Convolutional neural networks Decoding Encoding Encoding-Decoding Estimates lossless geometry compression octree coding Octrees Point cloud compression Process parameters Training Training data
ISSN:	2169-3536, 2169-3536
On-line prístup:	Získať plný text
Tagy:	Pridať tag Žiadne tagy, Buďte prvý, kto otaguje tento záznam!

Abstract	In this paper we propose a new paradigm for encoding the geometry of dense point cloud sequences, where a convolutional neural network (CNN), which estimates the encoding distributions, is optimized on several frames of the sequence to be compressed. We adopt lightweight CNN structures, we perform training as part of the encoding process and the CNN parameters are transmitted as part of the bitstream. The newly proposed encoding scheme operates on the octree representation for each point cloud, consecutively encoding each octree resolution level. At every octree resolution level, the voxel grid is traversed section-by-section (each section being perpendicular to a selected coordinate axis), and in each section, the occupancies of groups of two-by-two voxels are encoded at once in a single arithmetic coding operation. A context for the conditional encoding distribution is defined for each two-by-two group of voxels based on the information available about the occupancy of the neighboring voxels in the current and lower resolution layers of the octree. The CNN estimates the probability mass functions of the occupancy patterns of all the voxel groups from one section in four phases. In each new phase, the contexts are updated with the occupancies encoded in the previous phase, and each phase estimates the probabilities in parallel, providing a reasonable trade-off between the parallelism of the processing and the informativeness of the contexts. The CNN training time is comparable to the time spent in the remaining encoding steps, leading to competitive overall encoding times. The bitrates and encoding-decoding times compare favorably with those of recently published compression schemes.
AbstractList	In this paper we propose a new paradigm for encoding the geometry of dense point cloud sequences, where a convolutional neural network (CNN), which estimates the encoding distributions, is optimized on several frames of the sequence to be compressed. We adopt lightweight CNN structures, we perform training as part of the encoding process and the CNN parameters are transmitted as part of the bitstream. The newly proposed encoding scheme operates on the octree representation for each point cloud, consecutively encoding each octree resolution level. At every octree resolution level, the voxel grid is traversed section-by-section (each section being perpendicular to a selected coordinate axis), and in each section, the occupancies of groups of two-by-two voxels are encoded at once in a single arithmetic coding operation. A context for the conditional encoding distribution is defined for each two-by-two group of voxels based on the information available about the occupancy of the neighboring voxels in the current and lower resolution layers of the octree. The CNN estimates the probability mass functions of the occupancy patterns of all the voxel groups from one section in four phases. In each new phase, the contexts are updated with the occupancies encoded in the previous phase, and each phase estimates the probabilities in parallel, providing a reasonable trade-off between the parallelism of the processing and the informativeness of the contexts. The CNN training time is comparable to the time spent in the remaining encoding steps, leading to competitive overall encoding times. The bitrates and encoding-decoding times compare favorably with those of recently published compression schemes.
Author	Kaya, Emre C. Tabus, Ioan
Author_xml	– sequence: 1 givenname: Emre C. orcidid: 0000-0002-8255-1302 surname: Kaya fullname: Kaya, Emre C. email: emre.kaya@tuni.fi organization: Computing Sciences Unit, Tampere University, Tampere, Finland – sequence: 2 givenname: Ioan orcidid: 0000-0003-3131-9551 surname: Tabus fullname: Tabus, Ioan organization: Computing Sciences Unit, Tampere University, Tampere, Finland
BookMark	eNp9UU1PGzEQtRBIUMov4GKJc9KxvY7XR7SiLVL4qAJny2vPIkebdbA3h_Lr63QBIQ7MZcZP857G730jh0MckJBzBnPGQP-4bJqr1WrOgfO5YFpxLQ_ICWcLPRNSLA4_zMfkLOc1lKoLJNUJ-bOMOfeYM23iZpvKEOJAY0fvYxhG2vRx5-kKn3c4OMz0MYfh6f1N77Zj2IQX9LS5vaU30WOfv5OjzvYZz177KXn8efXQ_J4t735dN5fLmaugHmeOtbXkyivhRSc4dMAloOfOtozrSnrurS8IYNUiKCG043VbuVZ63TmQ4pRcT7o-2rXZprCx6a-JNpj_QExPxqYxuB5NJVqGrXJCLVylZKU7EAvnNCoJDHhdtC4mrW2K5Wt5NOu4S0M533AFArQGvd_S05ZLxbSEnXFhtGMxbEw29IaB2QdipkDMPhDzGkjhik_ct4u_Zp1PrICI7wxdjOMcxD9_dZeb
CODEN	IAECCG
CitedBy_id	crossref_primary_10_1109_ACCESS_2024_3469549
Cites_doi	10.1109/CVPR.2019.01088 10.1109/ICME46284.2020.9102866 10.1109/ICIP.2017.8296514 10.21437/Interspeech.2019-3174 10.1017/ATSIP.2020.12 10.1109/LSP.2018.2823701 10.1109/LSP.2020.2965322 10.1109/MMUL.2016.64 10.1109/ICIP.2017.8297036 10.1109/ICIP42928.2021.9506767 10.1109/ICIP40778.2020.9190648 10.1109/TIP.2020.3011811 10.1109/TCSVT.2021.3051377 10.1109/ICASSP.2019.8682413 10.1109/TIP.2019.2931466 10.1109/JSTSP.2020.3047520 10.1109/ICMEW53276.2021.9455990 10.3389/frsip.2022.846972 10.1109/ICRA.2012.6224647 10.1016/0146-664X(82)90104-6 10.1109/ICIP.2019.8803413 10.1109/ICCV.2015.123 10.1109/LSP.2021.3102525 10.1109/DCC50243.2021.00015 10.1109/TIP.2017.2707807 10.1109/TIP.2022.3185541 10.1109/ICIP.2018.8451802 10.1109/TCSVT.2021.3100279 10.1109/CVPR42600.2020.00139 10.1109/TCSVT.2016.2543039 10.1145/3343031.3351061 10.1109/ICASSP39728.2021.9414763 10.1109/ICASSP39728.2021.9413902 10.1109/MSP.2019.2900721 10.1109/DCC50243.2021.00014 10.1109/MMSP53017.2021.9733658 10.1109/ICIP42928.2021.9506429 10.1109/TBC.2019.2957652 10.1109/JETCAS.2018.2885981 10.1109/CVPR46437.2021.00598 10.1109/ICIP42928.2021.9506631 10.1109/JPROC.2021.3085957
ContentType	Journal Article
Copyright	Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2022
Copyright_xml	– notice: Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2022
DBID	97E ESBDL RIA RIE AAYXX CITATION 7SC 7SP 7SR 8BQ 8FD JG9 JQ2 L7M L~C L~D DOA
DOI	10.1109/ACCESS.2022.3197295
DatabaseName	IEEE Xplore (IEEE) IEEE Xplore Open Access Journals (WRLC) IEEE All-Society Periodicals Package (ASPP) 1998–Present IEEE Electronic Library (IEL) CrossRef Computer and Information Systems Abstracts Electronics & Communications Abstracts Engineered Materials Abstracts METADEX Technology Research Database Materials Research Database ProQuest Computer Science Collection Advanced Technologies Database with Aerospace Computer and Information Systems Abstracts Academic Computer and Information Systems Abstracts Professional DOAJ Directory of Open Access Journals
DatabaseTitle	CrossRef Materials Research Database Engineered Materials Abstracts Technology Research Database Computer and Information Systems Abstracts – Academic Electronics & Communications Abstracts ProQuest Computer Science Collection Computer and Information Systems Abstracts Advanced Technologies Database with Aerospace METADEX Computer and Information Systems Abstracts Professional
DatabaseTitleList	Materials Research Database
Database_xml	– sequence: 1 dbid: DOA name: DOAJ Directory of Open Access Journals url: https://www.doaj.org/ sourceTypes: Open Website – sequence: 2 dbid: RIE name: IEEE Electronic Library (IEL) url: https://ieeexplore.ieee.org/ sourceTypes: Publisher
DeliveryMethod	fulltext_linktorsrc
Discipline	Engineering
EISSN	2169-3536
EndPage	83691
ExternalDocumentID	oai_doaj_org_article_43b1eb7c376c47549f036cc9e7501028 10_1109_ACCESS_2022_3197295 9852220
Genre	orig-research
GroupedDBID	0R~ 4.4 5VS 6IK 97E AAJGR ABAZT ABVLG ACGFS ADBBV AGSQL ALMA_UNASSIGNED_HOLDINGS BCNDV BEFXN BFFAM BGNUA BKEBE BPEOZ EBS EJD ESBDL GROUPED_DOAJ IPLJI JAVBF KQ8 M43 M~E O9- OCL OK1 RIA RIE RNS AAYXX CITATION 7SC 7SP 7SR 8BQ 8FD JG9 JQ2 L7M L~C L~D
ID	FETCH-LOGICAL-c408t-c1b8527d73d3f320f0250ed2cab12945d2dad50e0e4be07339c28b4cb5d9fc053
IEDL.DBID	DOA
ISICitedReferencesCount	2
ISICitedReferencesURI	http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000842538600001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN	2169-3536
IngestDate	Fri Oct 03 12:52:08 EDT 2025 Mon Jun 30 03:44:04 EDT 2025 Tue Nov 18 21:48:02 EST 2025 Sat Nov 29 06:32:18 EST 2025 Wed Aug 27 02:22:59 EDT 2025
IsDoiOpenAccess	true
IsOpenAccess	true
IsPeerReviewed	true
IsScholarly	true
Language	English
License	https://creativecommons.org/licenses/by/4.0/legalcode
LinkModel	DirectLink
MergedId	FETCHMERGED-LOGICAL-c408t-c1b8527d73d3f320f0250ed2cab12945d2dad50e0e4be07339c28b4cb5d9fc053
Notes	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14
ORCID	0000-0002-8255-1302 0000-0003-3131-9551
OpenAccessLink	https://doaj.org/article/43b1eb7c376c47549f036cc9e7501028
PQID	2703099098
PQPubID	4845423
PageCount	14
ParticipantIDs	doaj_primary_oai_doaj_org_article_43b1eb7c376c47549f036cc9e7501028 proquest_journals_2703099098 crossref_citationtrail_10_1109_ACCESS_2022_3197295 crossref_primary_10_1109_ACCESS_2022_3197295 ieee_primary_9852220
PublicationCentury	2000
PublicationDate	20220000 2022-00-00 20220101 2022-01-01
PublicationDateYYYYMMDD	2022-01-01
PublicationDate_xml	– year: 2022 text: 20220000
PublicationDecade	2020
PublicationPlace	Piscataway
PublicationPlace_xml	– name: Piscataway
PublicationTitle	IEEE access
PublicationTitleAbbrev	Access
PublicationYear	2022
Publisher	IEEE The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Publisher_xml	– name: IEEE – name: The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
References	ref13 ref35 ref12 ref15 Maas (ref29); 30 ref37 ref14 ref31 d’Eon (ref34) 2017 ref30 ref11 ref10 ref32 ref2 ref1 ref17 ref39 ref16 ref38 ref19 ref18 Schwarz (ref36) 2018 (ref5) 2020 ref24 ref46 ref23 ref45 ref26 ref25 ref47 ref20 ref42 ref41 ref22 ref44 ref21 ref43 ref28 ref27 ref8 ref7 ref9 ref4 ref3 ref6 Loop (ref33) 2016 ref40
References_xml	– ident: ref35 doi: 10.1109/CVPR.2019.01088 – ident: ref22 doi: 10.1109/ICME46284.2020.9102866 – ident: ref15 doi: 10.1109/ICIP.2017.8296514 – ident: ref31 doi: 10.21437/Interspeech.2019-3174 – ident: ref4 doi: 10.1017/ATSIP.2020.12 – ident: ref19 doi: 10.1109/LSP.2018.2823701 – ident: ref6 doi: 10.1109/LSP.2020.2965322 – ident: ref2 doi: 10.1109/MMUL.2016.64 – ident: ref41 doi: 10.1109/ICIP.2017.8297036 – ident: ref13 doi: 10.1109/ICIP42928.2021.9506767 – ident: ref7 doi: 10.1109/ICIP40778.2020.9190648 – ident: ref10 doi: 10.1109/TIP.2020.3011811 – ident: ref21 doi: 10.1109/TCSVT.2021.3051377 – ident: ref47 doi: 10.1109/ICASSP.2019.8682413 – ident: ref18 doi: 10.1109/TIP.2019.2931466 – ident: ref27 doi: 10.1109/JSTSP.2020.3047520 – ident: ref26 doi: 10.1109/ICMEW53276.2021.9455990 – ident: ref25 doi: 10.3389/frsip.2022.846972 – ident: ref42 doi: 10.1109/ICRA.2012.6224647 – volume-title: TMC13 year: 2020 ident: ref5 – ident: ref28 doi: 10.1016/0146-664X(82)90104-6 – ident: ref14 doi: 10.1109/ICIP.2019.8803413 – ident: ref30 doi: 10.1109/ICCV.2015.123 – ident: ref12 doi: 10.1109/LSP.2021.3102525 – volume-title: Microsoft Voxelized Upper Bodies—A Voxelized Point Cloud Dataset year: 2016 ident: ref33 – ident: ref8 doi: 10.1109/DCC50243.2021.00015 – ident: ref39 doi: 10.1109/TIP.2017.2707807 – ident: ref32 doi: 10.1109/TIP.2022.3185541 – ident: ref40 doi: 10.1109/ICIP.2018.8451802 – ident: ref24 doi: 10.1109/TCSVT.2021.3100279 – ident: ref16 doi: 10.1109/CVPR42600.2020.00139 – ident: ref37 doi: 10.1109/TCSVT.2016.2543039 – volume-title: Common Test Conditions for Point Cloud Compression year: 2018 ident: ref36 – ident: ref44 doi: 10.1145/3343031.3351061 – ident: ref11 doi: 10.1109/ICASSP39728.2021.9414763 – ident: ref46 doi: 10.1109/ICASSP39728.2021.9413902 – ident: ref3 doi: 10.1109/MSP.2019.2900721 – ident: ref23 doi: 10.1109/DCC50243.2021.00014 – ident: ref17 doi: 10.1109/MMSP53017.2021.9733658 – ident: ref43 doi: 10.1109/ICIP42928.2021.9506429 – ident: ref9 doi: 10.1109/TBC.2019.2957652 – volume: 30 start-page: 3 volume-title: Proc. ICML ident: ref29 article-title: Rectifier nonlinearities improve neural network acoustic models – ident: ref1 doi: 10.1109/JETCAS.2018.2885981 – ident: ref20 doi: 10.1109/CVPR46437.2021.00598 – ident: ref45 doi: 10.1109/ICIP42928.2021.9506631 – volume-title: 8I Voxelized Full Bodies—A Voxelized Point Cloud Dataset year: 2017 ident: ref34 – ident: ref38 doi: 10.1109/JPROC.2021.3085957
SSID	ssj0000816957
Score	2.2494311
Snippet	In this paper we propose a new paradigm for encoding the geometry of dense point cloud sequences, where a convolutional neural network (CNN), which estimates...
SourceID	doaj proquest crossref ieee
SourceType	Open Website Aggregation Database Enrichment Source Index Database Publisher
StartPage	83678
SubjectTerms	Arithmetic coding Artificial neural networks Context modeling Convolutional neural networks Decoding Encoding Encoding-Decoding Estimates lossless geometry compression octree coding Octrees Point cloud compression Process parameters Training Training data
SummonAdditionalLinks	– databaseName: IEEE Electronic Library (IEL) dbid: RIE link: http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LT9wwELYo4tAeoEArFmjlQ48EYsdex0cagXpAW6oWxM2K7bG00jap2N0e-usZO96oqFWl3hLLjib5PM-MZwj5oHSQ3ntRtCBEISCURTvVvpDtlDFopbKpS8T9jZrN6ocHfbtFzsazMACQks_gPF6mf_m-d-sYKrvQNVoLHB30F0pNh7NaYzwlNpDQUuXCQqzUF5dNg--ALiDn6JlqtCLlM-WTavTnpip_SOKkXq73_o-w12Q3m5H0csB9n2xBd0Be_VZc8JB8uUESFijIaOT5Id21o32gt_28W9Fm0a89_brJpKYpd2C8p59RlHyf_wJPm9mMxo5pi-Ubcnd99a35VOQGCoUTZb0qHLNImPKq8lWoeBmiwQOeu9aimhfSc996HClBWIjdG7XjtRXOSq-DQ_Z8S7a7voMjQmuoAlPMSXTQhGRWW1v7tmKCCV9OQzkhfPNljcvVxWOTi4VJXkapzQCHiXCYDMeEnI2LfgzFNf49_WOEbJwaK2OnAcTCZEYzorIMrHIoOJ1Q6P0G1NHOaUDTKBpTE3IY8RsfkqGbkNPNBjCZi5eGR3GI6lrXx39fdUJeRgKHkMwp2V49ruEd2XE_V_Pl4_u0QZ8AywDiYw priority: 102 providerName: IEEE
Title	Lossless Compression of Point Cloud Sequences Using Sequence Optimized CNN Models
URI	https://ieeexplore.ieee.org/document/9852220 https://www.proquest.com/docview/2703099098 https://doaj.org/article/43b1eb7c376c47549f036cc9e7501028
Volume	10
WOSCitedRecordID	wos000842538600001&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
journalDatabaseRights	– providerCode: PRVAON databaseName: DOAJ Directory of Open Access Journals customDbUrl: eissn: 2169-3536 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0000816957 issn: 2169-3536 databaseCode: DOA dateStart: 20130101 isFulltext: true titleUrlDefault: https://www.doaj.org/ providerName: Directory of Open Access Journals – providerCode: PRVHPJ databaseName: ROAD: Directory of Open Access Scholarly Resources customDbUrl: eissn: 2169-3536 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0000816957 issn: 2169-3536 databaseCode: M~E dateStart: 20130101 isFulltext: true titleUrlDefault: https://road.issn.org providerName: ISSN International Centre
link	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwrV1LS8QwEA4iHvQgPnF9kYNHi0mabJqjFsWDrooPvIXmBQtrV3ZXDx787U7S7rIi6MVLoSFNk8nMN5ky_QahI6mCcM7xrPKcZ9wHklVd5TJRdSn1lZAmVYl4upK9XvH8rG7nSn3FnLCGHrgR3AnPDfVGWjAEyyVEMwEw11rlwdVF5xjRl0g1F0wlDC5oVwnZ0gxRok5OyxJWBAEhYxCnKjhTim-uKDH2tyVWfuBycjYXa2i1PSXi02Z262jB1xtoZY47cBPdXcGYA8ApHE26yWat8TDg22G_nuByMHxz-H6aKI1TasDsHt8AUrz0P7zDZa-HY0G0wXgLPV6cP5SXWVsfIbOcFJPMUlMIJp3MXR5yRkI8z3jHbGXAi3PhmKsctBDPjY_FGZVlheHWCKeCBevbRov1sPY7CBc-D1RSKyD-4oIaZUzhqpxyyh3pBtJBbCoqbVvy8FjDYqBTEEGUbuSro3x1K98OOp499NpwZ_ze_SzuwaxrJL5ODaAOulUH_Zc6dNBm3MHZIAokxBjMf3-6o7o10rFmEe3AG6ti9z9evYeW43Ka7zP7aHEyevMHaMm-T_rj0WHST7hef54fpr8MvwDoueaL
linkProvider	Directory of Open Access Journals
linkToHtml	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1Lb9QwEB5VBQk48CoVCwV84NjQ2LHX8bFEVEUsoYiCerPil7TSkqDuLgd-PWPHG4FASNwSy46cfJ6XM54P4KVUQTjneNF5zgvuQ1l0c-UK0c0p9Z2QJrFEfFnItq2vrtTFHhxPZ2G89yn5zL-Kl-lfvhvsNm6VnagavQWGAfqNyJwlxtNa045KpJBQQubSQrRUJ6dNg2-BQSBjGJsq9CPFb-YnVenPtCp_6OJkYM7u_d_U7sPd7EiS0xH5B7Dn-4dw55fyggfwcYFTWKEqI1Hqx4TXngyBXAzLfkOa1bB15NMul5qk7IHpnnxAZfJ1-cM70rQtiZxpq_Uj-Hz25rI5LzKFQmF5WW8KSw1OTDpZuSpUrAzR5fGO2c6goefCMdc5bCk9Nz7yNyrLasOtEU4FiwJ6CPv90PvHQGpfBSqpFRiicUGNMqZ2XUU55a6ch3IGbPdltc31xSPNxUqnOKNUeoRDRzh0hmMGx9Ogb2N5jX93fx0hm7rG2tipAbHQWdQ0rwz1RlpUnZZLjH8DWmlrlUfnKLpTMziI-E0PydDN4Gi3AHSW47VmUSGiwVb1k7-PegG3zi_fL_TibfvuKdyOkx03aI5gf3O99c_gpv2-Wa6vn6fF-hPC4OWu
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=article&rft.atitle=Lossless+Compression+of+Point+Cloud+Sequences+Using+Sequence+Optimized+CNN+Models&rft.jtitle=IEEE+access&rft.au=Kaya%2C+Emre+C.&rft.au=Tabus%2C+Ioan&rft.date=2022&rft.pub=IEEE&rft.eissn=2169-3536&rft.volume=10&rft.spage=83678&rft.epage=83691&rft_id=info:doi/10.1109%2FACCESS.2022.3197295&rft.externalDocID=9852220
thumbnail_l	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=2169-3536&client=summon
thumbnail_m	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=2169-3536&client=summon
thumbnail_s	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=2169-3536&client=summon