Second-order random graphs for modelling sets of attributed graphs and their application to object learning and recognition

The aim of this article is to present a random graph representation, that is based on second-order relations between graph elements, for modeling sets of attributed graphs (AGs). We refer to these models as Second-Order Random Graphs (SORGs). The basic feature of SORGs is that they include both marg...

Descripción completa

Detalles Bibliográficos
Autores: Sanfeliu, Alberto, Serratosa, Francesc, Alquézar Mancho, Renato
Tipo de recurso: artículo
Fecha de publicación:2004
País:España
Institución:Consejo Superior de Investigaciones Científicas (CSIC)
Repositorio:DIGITAL.CSIC. Repositorio Institucional del CSIC
OAI Identifier:oai:digital.csic.es:10261/30554
Acceso en línea:http://hdl.handle.net/10261/30554
Access Level:acceso abierto
Palabra clave:Random graphs
Graph synthesis
Distance measures
Object learning
Object recognition
Pattern recognition systems
id ES_4643adabe0145aa2b70469f7e238ff8d
oai_identifier_str oai:digital.csic.es:10261/30554
network_acronym_str ES
network_name_str España
repository_id_str
spelling Second-order random graphs for modelling sets of attributed graphs and their application to object learning and recognitionSanfeliu, AlbertoSerratosa, FrancescAlquézar Mancho, RenatoRandom graphsGraph synthesisDistance measuresObject learningObject recognitionPattern recognition systemsThe aim of this article is to present a random graph representation, that is based on second-order relations between graph elements, for modeling sets of attributed graphs (AGs). We refer to these models as Second-Order Random Graphs (SORGs). The basic feature of SORGs is that they include both marginal probability functions of graph elements and second-order joint probability functions. This allows a more precise description of both the structural and semantic information contents in a set of AGs and, consequently, an expected improvement in graph matching and object recognition. The article presents a probabilistic formulation of SORGs that includes as particular cases the two previously proposed approaches based on random graphs, namely the First-Order Random Graphs (FORGs) and the Function-Described Graphs (FDGs). We then propose a distance measure derived from the probability of instantiating a SORG into an AG and an incremental procedure to synthesize SORGs from sequences of AGs. Finally, SORGs are shown to improve the performance of FORGs, FDGs and direct AG-to-AG matching in three experimental recognition tasks: one in which AGs are randomly generated and the other two in which AGs represent multiple views of 3D objects (either synthetic or real) that have been extracted from color images. In the last case, object learning is achieved through the synthesis of SORG models.This work was supported by the project 'Active vision systems based in automatic learning for industrial applications' ().Peer ReviewedWorld Scientific Publishing201020102004info:eu-repo/semantics/articlehttp://purl.org/coar/resource_type/c_6501http://hdl.handle.net/10261/30554reponame:DIGITAL.CSIC. Repositorio Institucional del CSICinstname:Consejo Superior de Investigaciones Científicas (CSIC)Ingléshttp://dx.doi.org/10.1142/S0218001404003253info:eu-repo/semantics/openAccessoai:digital.csic.es:10261/305542026-05-22T06:33:51Z
dc.title.none.fl_str_mv Second-order random graphs for modelling sets of attributed graphs and their application to object learning and recognition
title Second-order random graphs for modelling sets of attributed graphs and their application to object learning and recognition
spellingShingle Second-order random graphs for modelling sets of attributed graphs and their application to object learning and recognition
Sanfeliu, Alberto
Random graphs
Graph synthesis
Distance measures
Object learning
Object recognition
Pattern recognition systems
title_short Second-order random graphs for modelling sets of attributed graphs and their application to object learning and recognition
title_full Second-order random graphs for modelling sets of attributed graphs and their application to object learning and recognition
title_fullStr Second-order random graphs for modelling sets of attributed graphs and their application to object learning and recognition
title_full_unstemmed Second-order random graphs for modelling sets of attributed graphs and their application to object learning and recognition
title_sort Second-order random graphs for modelling sets of attributed graphs and their application to object learning and recognition
dc.creator.none.fl_str_mv Sanfeliu, Alberto
Serratosa, Francesc
Alquézar Mancho, Renato
author Sanfeliu, Alberto
author_facet Sanfeliu, Alberto
Serratosa, Francesc
Alquézar Mancho, Renato
author_role author
author2 Serratosa, Francesc
Alquézar Mancho, Renato
author2_role author
author
dc.subject.none.fl_str_mv Random graphs
Graph synthesis
Distance measures
Object learning
Object recognition
Pattern recognition systems
topic Random graphs
Graph synthesis
Distance measures
Object learning
Object recognition
Pattern recognition systems
description The aim of this article is to present a random graph representation, that is based on second-order relations between graph elements, for modeling sets of attributed graphs (AGs). We refer to these models as Second-Order Random Graphs (SORGs). The basic feature of SORGs is that they include both marginal probability functions of graph elements and second-order joint probability functions. This allows a more precise description of both the structural and semantic information contents in a set of AGs and, consequently, an expected improvement in graph matching and object recognition. The article presents a probabilistic formulation of SORGs that includes as particular cases the two previously proposed approaches based on random graphs, namely the First-Order Random Graphs (FORGs) and the Function-Described Graphs (FDGs). We then propose a distance measure derived from the probability of instantiating a SORG into an AG and an incremental procedure to synthesize SORGs from sequences of AGs. Finally, SORGs are shown to improve the performance of FORGs, FDGs and direct AG-to-AG matching in three experimental recognition tasks: one in which AGs are randomly generated and the other two in which AGs represent multiple views of 3D objects (either synthetic or real) that have been extracted from color images. In the last case, object learning is achieved through the synthesis of SORG models.
publishDate 2004
dc.date.none.fl_str_mv 2004
2010
2010
dc.type.none.fl_str_mv info:eu-repo/semantics/article
http://purl.org/coar/resource_type/c_6501
format article
dc.identifier.none.fl_str_mv http://hdl.handle.net/10261/30554
url http://hdl.handle.net/10261/30554
dc.language.none.fl_str_mv Inglés
language_invalid_str_mv Inglés
dc.relation.none.fl_str_mv http://dx.doi.org/10.1142/S0218001404003253
dc.rights.none.fl_str_mv info:eu-repo/semantics/openAccess
eu_rights_str_mv openAccess
dc.publisher.none.fl_str_mv World Scientific Publishing
publisher.none.fl_str_mv World Scientific Publishing
dc.source.none.fl_str_mv reponame:DIGITAL.CSIC. Repositorio Institucional del CSIC
instname:Consejo Superior de Investigaciones Científicas (CSIC)
instname_str Consejo Superior de Investigaciones Científicas (CSIC)
reponame_str DIGITAL.CSIC. Repositorio Institucional del CSIC
collection DIGITAL.CSIC. Repositorio Institucional del CSIC
repository.name.fl_str_mv
repository.mail.fl_str_mv
_version_ 1869407216269787136
score 15,811543

Similares en LA Referencia