Logo do repositório
 
Publicação

Scalable Graph-Guided Transformer for Point Cloud Geometry Coding

2025Artigo de investigação Acesso restrito
Ver registo simples
datacite.subject.fosEngenharia e Tecnologia::Engenharia Eletrotécnica, Eletrónica e Informática
datacite.subject.sdg09:Indústria, Inovação e Infraestruturas
dc.contributor.authorGhafari, Mohammadreza
dc.contributor.authorGuarda, André F. R.
dc.contributor.authorRodrigues, Nuno M. M.
dc.contributor.authorPereira, Fernando
dc.date.accessioned2025-12-12T16:07:45Z
dc.date.available2025-12-12T16:07:45Z
dc.date.issued2025en_US
dc.date.updated2025-12-12T13:24:48Z
dc.description
dc.description.abstractAttention models, particularly Transformers, have significantly advanced deep learning in fields like natural language processing and computer vision by capturing contextual relationships in both sequential and spatial data. This ability is valuable for Point Clouds (PC), which are unstructured sets of points in 3D space. Transformers can effectively identify correlations between distant points, allowing them to focus on the most critical regions of the data. To demonstrate this capability, this paper proposes a novel, scalable Graph-Guided Transformer model, labeled 2GFormer, for static PC geometry. This model is built using a scalable architecture that leverages Graph Convolutions to enhance a Relational Neighborhood SelfAttention (RNSA) base layer model. Both models are integrated into the JPEG Pleno Learning-based Point Cloud Coding (JPEG PCC) standard, resulting in the creation of two attention-enabled codecs for static PC coding: JPEG RNSA and JPEG 2GFormer. While JPEG RNSA codec delivers significant compression improvements for solid and dense PCs compared to the baseline JPEG PCC standard, JPEG 2GFormer extends these gains to solid, dense, and sparse PCs with only a marginal increase in model parameters. Additionally, JPEG 2GFormer outperforms both conventional and learning-based state-of-the-art PC codecs. These results position JPEG 2GFormer as a highly efficient solution for versatile PC coding.eng
dc.description.sponsorshipThis work was funded by the Fundação para a Ciência e a Tecnologia (FCT, Portugal) through the research project PTDC/EEI-COM/1125/2021, entitled “Deep Learning-based Point Cloud Representation.”
dc.description.versionN/A
dc.identifier.citationM. Ghafari, A. F. R. Guarda, N. M. M. Rodrigues and F. Pereira, "Scalable Graph-Guided Transformer for Point Cloud Geometry Coding," in IEEE Transactions on Multimedia, doi: 10.1109/TMM.2025.3598605.
dc.identifier.doi10.1109/tmm.2025.3598605en_US
dc.identifier.issn1520-9210en_US
dc.identifier.issn1941-0077en_US
dc.identifier.slugcv-prod-4622938
dc.identifier.urihttp://hdl.handle.net/10400.8/15021
dc.language.isoeng
dc.peerreviewedyes
dc.publisherIEEE
dc.relationDeep learning-based Point Cloud Representation
dc.relation.hasversionhttps://ieeexplore.ieee.org/document/11123804
dc.rights.urihttp://creativecommons.org/licenses/by/4.0/
dc.subjectGraph Convolutions
dc.subjectJPEG Pleno
dc.subjectPoint Cloud Coding
dc.subjectScalable Transformer
dc.subjectSelf-Attention
dc.titleScalable Graph-Guided Transformer for Point Cloud Geometry Codingeng
dc.typeresearch articleen_US
dspace.entity.typePublication
oaire.awardTitleDeep learning-based Point Cloud Representation
oaire.awardURIinfo:eu-repo/grantAgreement/FCT/3599-PPCDT/PTDC%2FEEI-COM%2F1125%2F2021/PT
oaire.citation.endPage14
oaire.citation.startPage1
oaire.citation.titleIEEE Transactions on Multimediaen_US
oaire.fundingStream3599-PPCDT
oaire.versionhttp://purl.org/coar/version/c_ab4af688f83e57aa
person.familyNameGuarda
person.familyNameM. M. Rodrigues
person.givenNameAndré
person.givenNameNuno
person.identifier.ciencia-idF811-146F-4EE9
person.identifier.ciencia-id6917-B121-4E34
person.identifier.orcid0000-0001-5996-1074
person.identifier.orcid0000-0001-9536-1017
person.identifier.scopus-author-id7006052345
project.funder.identifierhttp://doi.org/10.13039/501100001871
project.funder.nameFundação para a Ciência e a Tecnologia
rcaap.cv.cienciaid6917-B121-4E34 | NUNO MIGUEL MORAIS RODRIGUES
rcaap.rightsclosedAccessen_US
relation.isAuthorOfPublicationab4d7e6e-b391-49ba-a618-a52fc62c8837
relation.isAuthorOfPublicationb4ebe652-7f0e-4e67-adb0-d5ea29fc9e69
relation.isAuthorOfPublication.latestForDiscoveryb4ebe652-7f0e-4e67-adb0-d5ea29fc9e69
relation.isProjectOfPublicationa018adbc-131c-448f-acdc-fad90c525470
relation.isProjectOfPublication.latestForDiscoverya018adbc-131c-448f-acdc-fad90c525470

Ficheiros

Principais
A mostrar 1 - 2 de 2
Miniatura indisponível
Nome:
FINAL_VERSION.pdf
Tamanho:
1.19 MB
Formato:
Adobe Portable Document Format
Ver/Abrir
Miniatura
Nome:
ScalableGraph-GuidedTransformerforPointCloud_Accepted_version.pdf
Tamanho:
3.31 MB
Formato:
Adobe Portable Document Format
Descrição:
This article has been accepted for publication in IEEE Transactions on Multimedia. This is the author's version which has not been fully edited and content may change prior to final publication.
Ver/Abrir
Licença
A mostrar 1 - 1 de 1
Miniatura indisponível
Nome:
license.txt
Tamanho:
1.33 KB
Formato:
Item-specific license agreed upon to submission
Descrição:
Ver/Abrir

Coleções

ESTG - Artigos em revistas internacionais