Motion Textures: Modeling, Classification And Segmentation Using Mixed-State Markov Random Fields

Autores
Yao, Jian-Feng; Crivelli, Tomás; Cernuschi Frias, Bruno; Bouthemy, P.
Año de publicación
2013
Idioma
inglés
Tipo de recurso
artículo
Estado
versión publicada
Descripción
A motion texture is an instantaneous motion map extracted from a dynamic texture. We observe that such motion maps exhibit values of two types: a discrete component at zero (absence of motion) and continuous motion values. We thus develop a mixed-state Markov random field model to represent motion textures. The core of our approach is to show that motion information is powerful enough to classify and segment dynamic textures if it is properly modeled regarding its specific nature and the local interactions involved. A parsimonious set of 11 parameters constitutes the descriptive feature of a motion texture. The motivation of the proposed formulation runs toward the analysis of dynamic video contents, and we tackle two related problems. First, we present a method for recognition and classification of motion textures, by means of the Kullback-Leibler distance between mixed-state statistical models. Second, we define a two-frame motion texture maximum a posteriori (MAP)-based segmentation method applicable to motion textures with deforming boundaries. We also investigate a new issue, the space-time dynamic texture segmentation, by combining the spatial segmentation and the recognition methods. Numerous experimental results are reported for those three problems which demonstrate the efficiency and accuracy of the proposed two-frame approach.
Fil: Yao, Jian-Feng. The University of Hong Kong. Department of Statistics and Actuarial Science; Hong Kong
Fil: Crivelli, Tomás. Universidad de Buenos Aires. Facultad de Ingenieria. Departamento de Electronica; Argentina. Consejo Nacional de Investigaciones Científicas y Técnicas; Argentina
Fil: Cernuschi Frias, Bruno. Consejo Nacional de Investigaciones Científicas y Técnicas. Oficina de Coordinación Administrativa Saavedra 15. Instituto Argentino de Matemáticas; Argentina
Fil: Bouthemy, P.. Inria, Centre Rennes - Bretagne Atlantique; Francia
Materia
Dynamic Textures
Random Fields
Motion Analysis
Mixed-State Models
Nivel de accesibilidad
acceso abierto
Condiciones de uso
https://creativecommons.org/licenses/by-nc-nd/2.5/ar/
Repositorio
CONICET Digital (CONICET)
Institución
Consejo Nacional de Investigaciones Científicas y Técnicas
OAI Identificador
oai:ri.conicet.gov.ar:11336/3377
Acceder
id CONICETDig_73b46969fe2938689d459540e8668185
oai_identifier_str oai:ri.conicet.gov.ar:11336/3377
network_acronym_str CONICETDig
repository_id_str 3498
network_name_str CONICET Digital (CONICET)
spelling Motion Textures: Modeling, Classification And Segmentation Using Mixed-State Markov Random FieldsYao, Jian-FengCrivelli, TomásCernuschi Frias, BrunoBouthemy, P.Dynamic TexturesRandom FieldsMotion AnalysisMixed-State Modelshttps://purl.org/becyt/ford/2.2https://purl.org/becyt/ford/2A motion texture is an instantaneous motion map extracted from a dynamic texture. We observe that such motion maps exhibit values of two types: a discrete component at zero (absence of motion) and continuous motion values. We thus develop a mixed-state Markov random field model to represent motion textures. The core of our approach is to show that motion information is powerful enough to classify and segment dynamic textures if it is properly modeled regarding its specific nature and the local interactions involved. A parsimonious set of 11 parameters constitutes the descriptive feature of a motion texture. The motivation of the proposed formulation runs toward the analysis of dynamic video contents, and we tackle two related problems. First, we present a method for recognition and classification of motion textures, by means of the Kullback-Leibler distance between mixed-state statistical models. Second, we define a two-frame motion texture maximum a posteriori (MAP)-based segmentation method applicable to motion textures with deforming boundaries. We also investigate a new issue, the space-time dynamic texture segmentation, by combining the spatial segmentation and the recognition methods. Numerous experimental results are reported for those three problems which demonstrate the efficiency and accuracy of the proposed two-frame approach.Fil: Yao, Jian-Feng. The University of Hong Kong. Department of Statistics and Actuarial Science; Hong KongFil: Crivelli, Tomás. Universidad de Buenos Aires. Facultad de Ingenieria. Departamento de Electronica; Argentina. Consejo Nacional de Investigaciones Científicas y Técnicas; ArgentinaFil: Cernuschi Frias, Bruno. Consejo Nacional de Investigaciones Científicas y Técnicas. Oficina de Coordinación Administrativa Saavedra 15. Instituto Argentino de Matemáticas; ArgentinaFil: Bouthemy, P.. Inria, Centre Rennes - Bretagne Atlantique; FranciaSIAM2013-12info:eu-repo/semantics/articleinfo:eu-repo/semantics/publishedVersionhttp://purl.org/coar/resource_type/c_6501info:ar-repo/semantics/articuloapplication/pdfapplication/pdfhttp://hdl.handle.net/11336/3377Yao, Jian-Feng ; Crivelli, Tomás; Cernuschi Frias, Bruno; Bouthemy, P.; Motion Textures: Modeling, Classification And Segmentation Using Mixed-State Markov Random Fields; SIAM; SIAM Journal On Imaging Sciences; 6; 4; 12-2013; 2484-25201936-4954enginfo:eu-repo/semantics/altIdentifier/doi/10.1137/120872048info:eu-repo/semantics/openAccesshttps://creativecommons.org/licenses/by-nc-nd/2.5/ar/reponame:CONICET Digital (CONICET)instname:Consejo Nacional de Investigaciones Científicas y Técnicas2025-09-29T10:21:01Zoai:ri.conicet.gov.ar:11336/3377instacron:CONICETInstitucionalhttp://ri.conicet.gov.ar/Organismo científico-tecnológicoNo correspondehttp://ri.conicet.gov.ar/oai/requestdasensio@conicet.gov.ar; lcarlino@conicet.gov.arArgentinaNo correspondeNo correspondeNo correspondeopendoar:34982025-09-29 10:21:01.662CONICET Digital (CONICET) - Consejo Nacional de Investigaciones Científicas y Técnicasfalse
dc.title.none.fl_str_mv Motion Textures: Modeling, Classification And Segmentation Using Mixed-State Markov Random Fields
title Motion Textures: Modeling, Classification And Segmentation Using Mixed-State Markov Random Fields
spellingShingle Motion Textures: Modeling, Classification And Segmentation Using Mixed-State Markov Random Fields
Yao, Jian-Feng
Dynamic Textures
Random Fields
Motion Analysis
Mixed-State Models
title_short Motion Textures: Modeling, Classification And Segmentation Using Mixed-State Markov Random Fields
title_full Motion Textures: Modeling, Classification And Segmentation Using Mixed-State Markov Random Fields
title_fullStr Motion Textures: Modeling, Classification And Segmentation Using Mixed-State Markov Random Fields
title_full_unstemmed Motion Textures: Modeling, Classification And Segmentation Using Mixed-State Markov Random Fields
title_sort Motion Textures: Modeling, Classification And Segmentation Using Mixed-State Markov Random Fields
dc.creator.none.fl_str_mv Yao, Jian-Feng
Crivelli, Tomás
Cernuschi Frias, Bruno
Bouthemy, P.
author Yao, Jian-Feng
author_facet Yao, Jian-Feng
Crivelli, Tomás
Cernuschi Frias, Bruno
Bouthemy, P.
author_role author
author2 Crivelli, Tomás
Cernuschi Frias, Bruno
Bouthemy, P.
author2_role author
author
author
dc.subject.none.fl_str_mv Dynamic Textures
Random Fields
Motion Analysis
Mixed-State Models
topic Dynamic Textures
Random Fields
Motion Analysis
Mixed-State Models
purl_subject.fl_str_mv https://purl.org/becyt/ford/2.2
https://purl.org/becyt/ford/2
dc.description.none.fl_txt_mv A motion texture is an instantaneous motion map extracted from a dynamic texture. We observe that such motion maps exhibit values of two types: a discrete component at zero (absence of motion) and continuous motion values. We thus develop a mixed-state Markov random field model to represent motion textures. The core of our approach is to show that motion information is powerful enough to classify and segment dynamic textures if it is properly modeled regarding its specific nature and the local interactions involved. A parsimonious set of 11 parameters constitutes the descriptive feature of a motion texture. The motivation of the proposed formulation runs toward the analysis of dynamic video contents, and we tackle two related problems. First, we present a method for recognition and classification of motion textures, by means of the Kullback-Leibler distance between mixed-state statistical models. Second, we define a two-frame motion texture maximum a posteriori (MAP)-based segmentation method applicable to motion textures with deforming boundaries. We also investigate a new issue, the space-time dynamic texture segmentation, by combining the spatial segmentation and the recognition methods. Numerous experimental results are reported for those three problems which demonstrate the efficiency and accuracy of the proposed two-frame approach.
Fil: Yao, Jian-Feng. The University of Hong Kong. Department of Statistics and Actuarial Science; Hong Kong
Fil: Crivelli, Tomás. Universidad de Buenos Aires. Facultad de Ingenieria. Departamento de Electronica; Argentina. Consejo Nacional de Investigaciones Científicas y Técnicas; Argentina
Fil: Cernuschi Frias, Bruno. Consejo Nacional de Investigaciones Científicas y Técnicas. Oficina de Coordinación Administrativa Saavedra 15. Instituto Argentino de Matemáticas; Argentina
Fil: Bouthemy, P.. Inria, Centre Rennes - Bretagne Atlantique; Francia
description A motion texture is an instantaneous motion map extracted from a dynamic texture. We observe that such motion maps exhibit values of two types: a discrete component at zero (absence of motion) and continuous motion values. We thus develop a mixed-state Markov random field model to represent motion textures. The core of our approach is to show that motion information is powerful enough to classify and segment dynamic textures if it is properly modeled regarding its specific nature and the local interactions involved. A parsimonious set of 11 parameters constitutes the descriptive feature of a motion texture. The motivation of the proposed formulation runs toward the analysis of dynamic video contents, and we tackle two related problems. First, we present a method for recognition and classification of motion textures, by means of the Kullback-Leibler distance between mixed-state statistical models. Second, we define a two-frame motion texture maximum a posteriori (MAP)-based segmentation method applicable to motion textures with deforming boundaries. We also investigate a new issue, the space-time dynamic texture segmentation, by combining the spatial segmentation and the recognition methods. Numerous experimental results are reported for those three problems which demonstrate the efficiency and accuracy of the proposed two-frame approach.
publishDate 2013
dc.date.none.fl_str_mv 2013-12
dc.type.none.fl_str_mv info:eu-repo/semantics/article
info:eu-repo/semantics/publishedVersion
http://purl.org/coar/resource_type/c_6501
info:ar-repo/semantics/articulo
format article
status_str publishedVersion
dc.identifier.none.fl_str_mv http://hdl.handle.net/11336/3377
Yao, Jian-Feng ; Crivelli, Tomás; Cernuschi Frias, Bruno; Bouthemy, P.; Motion Textures: Modeling, Classification And Segmentation Using Mixed-State Markov Random Fields; SIAM; SIAM Journal On Imaging Sciences; 6; 4; 12-2013; 2484-2520
1936-4954
url http://hdl.handle.net/11336/3377
identifier_str_mv Yao, Jian-Feng ; Crivelli, Tomás; Cernuschi Frias, Bruno; Bouthemy, P.; Motion Textures: Modeling, Classification And Segmentation Using Mixed-State Markov Random Fields; SIAM; SIAM Journal On Imaging Sciences; 6; 4; 12-2013; 2484-2520
1936-4954
dc.language.none.fl_str_mv eng
language eng
dc.relation.none.fl_str_mv info:eu-repo/semantics/altIdentifier/doi/10.1137/120872048
dc.rights.none.fl_str_mv info:eu-repo/semantics/openAccess
https://creativecommons.org/licenses/by-nc-nd/2.5/ar/
eu_rights_str_mv openAccess
rights_invalid_str_mv https://creativecommons.org/licenses/by-nc-nd/2.5/ar/
dc.format.none.fl_str_mv application/pdf
application/pdf
dc.publisher.none.fl_str_mv SIAM
publisher.none.fl_str_mv SIAM
dc.source.none.fl_str_mv reponame:CONICET Digital (CONICET)
instname:Consejo Nacional de Investigaciones Científicas y Técnicas
reponame_str CONICET Digital (CONICET)
collection CONICET Digital (CONICET)
instname_str Consejo Nacional de Investigaciones Científicas y Técnicas
repository.name.fl_str_mv CONICET Digital (CONICET) - Consejo Nacional de Investigaciones Científicas y Técnicas
repository.mail.fl_str_mv dasensio@conicet.gov.ar; lcarlino@conicet.gov.ar
_version_ 1844614196283572224
score 13.070432

Publicaciones similares