AustralieFrV1, PascalFrancis, Curation, bibRecord, 004773

Graph attribute embedding via Riemannian submersion learning

Identifieur interne : 004773 ( PascalFrancis/Curation ); précédent : 004772; suivant : 004774

Graph attribute embedding via Riemannian submersion learning

Auteurs : HAIFENG ZHAO [République populaire de Chine] ; Antonio Robles-Kelly [Australie] ; JUN ZHOU [Australie] ; JIANFENG LU [République populaire de Chine] ; Jing-Yu Yang [République populaire de Chine]

Source :

Computer vision and image understanding : (Print) [ 1077-3142 ] ; 2011.

RBID : Pascal:12-0017653

Descripteurs français

Pascal (Inist)
- Vision ordinateur, Concordance forme, Traitement image, Classification, Base de données, Espace métrique, Catégorisation, Estimation densité, Mesure densité, Intervalle confiance, Densité probabilité, Méthode itérative, Approximation L2, Sommet graphe, Probabilité a posteriori, Couplage graphe, Loi a posteriori, ., Reconnaissance objet.
Wicri :
- topic : Classification, Base de données.

English descriptors

KwdEn :
- Categorization, Classification, Computer vision, Confidence interval, Database, Density estimation, Density measurement, Graph matching, Image processing, Iterative method, L2 approximation, Metric space, Object recognition, Pattern matching, Posterior distribution, Posterior probability, Probability density, Vertex(graph).

Abstract

In this paper, we tackle the problem of embedding a set of relational structures into a metric space for purposes of matching and categorisation. To this end, we view the problem from a Riemannian perspective and make use of the concepts of charts on the manifold to define the embedding as a mixture of class-specific submersions. Formulated in this manner, the mixture weights are recovered using a probability density estimation on the embedded graph node coordinates. Further, we recover these class-specific submersions making use of an iterative trust-region method so as to minimise the L2 norm between the hard limit of the graph-vertex posterior probabilities and their estimated values. The method presented here is quite general in nature and allows tasks such as matching, categorisation and retrieval. We show results on graph matching, shape categorisation and digit classification on synthetic data, the MNIST dataset and the MPEG-7 database.

A01	`01`	`1`		`@0 1077-3142`
A02	`01`			`@0 CVIUF4`
A03		`1`		`@0 Comput. vis. image underst. : (Print)`
A05				`@2 115`
A06				`@2 7`
A08	`01`	`1`	`ENG`	`@1 Graph attribute embedding via Riemannian submersion learning`
A09	`01`	`1`	`ENG`	`@1 Special issue on Graph-Based Representations in Computer Vision`
A11	`01`	`1`		`@1 HAIFENG ZHAO`
A11	`02`	`1`		`@1 ROBLES-KELLY (Antonio)`
A11	`03`	`1`		`@1 JUN ZHOU`
A11	`04`	`1`		`@1 JIANFENG LU`
A11	`05`	`1`		`@1 YANG (Jing-Yu)`
A12	`01`	`1`		`@1 HANCOCK (Edwin R.) @9 ed.`
A12	`02`	`1`		`@1 TORSELLO (Andrea) @9 ed.`
A12	`03`	`1`		`@1 ESCOLANO (Francisco) @9 ed.`
A12	`04`	`1`		`@1 BRUN (Luc) @9 ed.`
A14	`01`			`@1 Tower A 7 London Circuit @2 Canberra ACT 2601 @3 AUS @Z 2 aut. @Z 3 aut.`
A14	`02`			`@1 College of Engineering and Computer Science, The Australian National University @2 Canberra ACT 0200 @3 AUS @Z 2 aut. @Z 3 aut.`
A14	`03`			`@1 UNSW@ADFA @2 Canberra ACT 2600 @3 AUS @Z 2 aut. @Z 3 aut.`
A14	`04`			`@1 Nanjing University of Science and Technology, 200 Xiaolingwei Street @2 Nanjing 210094 @3 CHN @Z 1 aut. @Z 4 aut. @Z 5 aut.`
A15	`01`			`@1 University of York @3 GBR @Z 1 aut.`
A15	`02`			`@1 Università Ca' Foscari Venezia @3 ITA @Z 2 aut.`
A15	`03`			`@1 University of Alicante @3 ESP @Z 3 aut.`
A15	`04`			`@1 GREYC, CNRS UMR 6072, ENSICAEN @3 FRA @Z 4 aut.`
A20				`@1 962-975`
A21				`@1 2011`
A23	`01`			`@0 ENG`
A43	`01`			`@1 INIST @2 15463A @5 354000509422430050`
A44				`@0 0000 @1 © 2012 INIST-CNRS. All rights reserved.`
A45				`@0 66 ref.`
A47	`01`	`1`		`@0 12-0017653`
A60				`@1 P`
A61				`@0 A`
A64	`01`	`1`		`@0 Computer vision and image understanding : (Print)`
A66	`01`			`@0 NLD`
C01	`01`		`ENG`	@0 In this paper, we tackle the problem of embedding a set of relational structures into a metric space for purposes of matching and categorisation. To this end, we view the problem from a Riemannian perspective and make use of the concepts of charts on the manifold to define the embedding as a mixture of class-specific submersions. Formulated in this manner, the mixture weights are recovered using a probability density estimation on the embedded graph node coordinates. Further, we recover these class-specific submersions making use of an iterative trust-region method so as to minimise the L2 norm between the hard limit of the graph-vertex posterior probabilities and their estimated values. The method presented here is quite general in nature and allows tasks such as matching, categorisation and retrieval. We show results on graph matching, shape categorisation and digit classification on synthetic data, the MNIST dataset and the MPEG-7 database.
C02	`01`	`X`		`@0 001D02A06`
C02	`02`	`X`		`@0 001D02C03`
C03	`01`	`X`	`FRE`	`@0 Vision ordinateur @5 01`
C03	`01`	`X`	`ENG`	`@0 Computer vision @5 01`
C03	`01`	`X`	`SPA`	`@0 Visión ordenador @5 01`
C03	`02`	`X`	`FRE`	`@0 Concordance forme @5 06`
C03	`02`	`X`	`ENG`	`@0 Pattern matching @5 06`
C03	`03`	`X`	`FRE`	`@0 Traitement image @5 07`
C03	`03`	`X`	`ENG`	`@0 Image processing @5 07`
C03	`03`	`X`	`SPA`	`@0 Procesamiento imagen @5 07`
C03	`04`	`X`	`FRE`	`@0 Classification @5 08`
C03	`04`	`X`	`ENG`	`@0 Classification @5 08`
C03	`04`	`X`	`SPA`	`@0 Clasificación @5 08`
C03	`05`	`X`	`FRE`	`@0 Base de données @5 09`
C03	`05`	`X`	`ENG`	`@0 Database @5 09`
C03	`05`	`X`	`SPA`	`@0 Base dato @5 09`
C03	`06`	`X`	`FRE`	`@0 Espace métrique @5 18`
C03	`06`	`X`	`ENG`	`@0 Metric space @5 18`
C03	`06`	`X`	`SPA`	`@0 Espacio métrico @5 18`
C03	`07`	`X`	`FRE`	`@0 Catégorisation @5 19`
C03	`07`	`X`	`ENG`	`@0 Categorization @5 19`
C03	`07`	`X`	`SPA`	`@0 Categorización @5 19`
C03	`08`	`X`	`FRE`	`@0 Estimation densité @5 20`
C03	`08`	`X`	`ENG`	`@0 Density estimation @5 20`
C03	`08`	`X`	`SPA`	`@0 Estimación densidad @5 20`
C03	`09`	`X`	`FRE`	`@0 Mesure densité @5 21`
C03	`09`	`X`	`ENG`	`@0 Density measurement @5 21`
C03	`09`	`X`	`SPA`	`@0 Medición densidad @5 21`
C03	`10`	`X`	`FRE`	`@0 Intervalle confiance @5 22`
C03	`10`	`X`	`ENG`	`@0 Confidence interval @5 22`
C03	`10`	`X`	`SPA`	`@0 Intervalo confianza @5 22`
C03	`11`	`X`	`FRE`	`@0 Densité probabilité @5 23`
C03	`11`	`X`	`ENG`	`@0 Probability density @5 23`
C03	`11`	`X`	`SPA`	`@0 Densidad probabilidad @5 23`
C03	`12`	`X`	`FRE`	`@0 Méthode itérative @5 24`
C03	`12`	`X`	`ENG`	`@0 Iterative method @5 24`
C03	`12`	`X`	`SPA`	`@0 Método iterativo @5 24`
C03	`13`	`X`	`FRE`	`@0 Approximation L2 @5 25`
C03	`13`	`X`	`ENG`	`@0 L2 approximation @5 25`
C03	`13`	`X`	`SPA`	`@0 Aproximación L2 @5 25`
C03	`14`	`X`	`FRE`	`@0 Sommet graphe @5 26`
C03	`14`	`X`	`ENG`	`@0 Vertex(graph) @5 26`
C03	`14`	`X`	`SPA`	`@0 Vértice grafo @5 26`
C03	`15`	`X`	`FRE`	`@0 Probabilité a posteriori @5 27`
C03	`15`	`X`	`ENG`	`@0 Posterior probability @5 27`
C03	`15`	`X`	`SPA`	`@0 Probabilidad a posteriori @5 27`
C03	`16`	`X`	`FRE`	`@0 Couplage graphe @5 28`
C03	`16`	`X`	`ENG`	`@0 Graph matching @5 28`
C03	`16`	`X`	`SPA`	`@0 Acoplamiento grafo @5 28`
C03	`17`	`X`	`FRE`	`@0 Loi a posteriori @5 41`
C03	`17`	`X`	`ENG`	`@0 Posterior distribution @5 41`
C03	`17`	`X`	`SPA`	`@0 Ley a posteriori @5 41`
C03	`18`	`X`	`FRE`	`@0 . @4 INC @5 82`
C03	`19`	`X`	`FRE`	`@0 Reconnaissance objet @4 CD @5 96`
C03	`19`	`X`	`ENG`	`@0 Object recognition @4 CD @5 96`
C03	`19`	`X`	`SPA`	`@0 Reconocimiento de objetos @4 CD @5 96`
N21				`@1 002`
N44	`01`			`@1 OTO`
N82				`@1 OTO`

Links toward previous steps (curation, corpus...)

to stream PascalFrancis, to step Corpus: Pour aller vers cette notice dans l'étape Curation :001736

Links to Exploration step

Pascal:12-0017653

Le document en format XML

<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en" level="a">Graph attribute embedding via Riemannian submersion learning</title>
<author><name sortKey="Haifeng Zhao" sort="Haifeng Zhao" uniqKey="Haifeng Zhao" last="Haifeng Zhao">HAIFENG ZHAO</name>
<affiliation wicri:level="1"><inist:fA14 i1="04"><s1>Nanjing University of Science and Technology, 200 Xiaolingwei Street</s1>
<s2>Nanjing 210094</s2>
<s3>CHN</s3>
<sZ>1 aut.</sZ>
<sZ>4 aut.</sZ>
<sZ>5 aut.</sZ>
</inist:fA14>
<country>République populaire de Chine</country>
</affiliation>
</author>
<author><name sortKey="Robles Kelly, Antonio" sort="Robles Kelly, Antonio" uniqKey="Robles Kelly A" first="Antonio" last="Robles-Kelly">Antonio Robles-Kelly</name>
<affiliation wicri:level="1"><inist:fA14 i1="01"><s1>Tower A 7 London Circuit</s1>
<s2>Canberra ACT 2601</s2>
<s3>AUS</s3>
<sZ>2 aut.</sZ>
<sZ>3 aut.</sZ>
</inist:fA14>
<country>Australie</country>
</affiliation>
<affiliation wicri:level="1"><inist:fA14 i1="02"><s1>College of Engineering and Computer Science, The Australian National University</s1>
<s2>Canberra ACT 0200</s2>
<s3>AUS</s3>
<sZ>2 aut.</sZ>
<sZ>3 aut.</sZ>
</inist:fA14>
<country>Australie</country>
</affiliation>
<affiliation wicri:level="1"><inist:fA14 i1="03"><s1>UNSW@ADFA</s1>
<s2>Canberra ACT 2600</s2>
<s3>AUS</s3>
<sZ>2 aut.</sZ>
<sZ>3 aut.</sZ>
</inist:fA14>
<country>Australie</country>
</affiliation>
</author>
<author><name sortKey="Jun Zhou" sort="Jun Zhou" uniqKey="Jun Zhou" last="Jun Zhou">JUN ZHOU</name>
<affiliation wicri:level="1"><inist:fA14 i1="01"><s1>Tower A 7 London Circuit</s1>
<s2>Canberra ACT 2601</s2>
<s3>AUS</s3>
<sZ>2 aut.</sZ>
<sZ>3 aut.</sZ>
</inist:fA14>
<country>Australie</country>
</affiliation>
<affiliation wicri:level="1"><inist:fA14 i1="02"><s1>College of Engineering and Computer Science, The Australian National University</s1>
<s2>Canberra ACT 0200</s2>
<s3>AUS</s3>
<sZ>2 aut.</sZ>
<sZ>3 aut.</sZ>
</inist:fA14>
<country>Australie</country>
</affiliation>
<affiliation wicri:level="1"><inist:fA14 i1="03"><s1>UNSW@ADFA</s1>
<s2>Canberra ACT 2600</s2>
<s3>AUS</s3>
<sZ>2 aut.</sZ>
<sZ>3 aut.</sZ>
</inist:fA14>
<country>Australie</country>
</affiliation>
</author>
<author><name sortKey="Jianfeng Lu" sort="Jianfeng Lu" uniqKey="Jianfeng Lu" last="Jianfeng Lu">JIANFENG LU</name>
<affiliation wicri:level="1"><inist:fA14 i1="04"><s1>Nanjing University of Science and Technology, 200 Xiaolingwei Street</s1>
<s2>Nanjing 210094</s2>
<s3>CHN</s3>
<sZ>1 aut.</sZ>
<sZ>4 aut.</sZ>
<sZ>5 aut.</sZ>
</inist:fA14>
<country>République populaire de Chine</country>
</affiliation>
</author>
<author><name sortKey="Yang, Jing Yu" sort="Yang, Jing Yu" uniqKey="Yang J" first="Jing-Yu" last="Yang">Jing-Yu Yang</name>
<affiliation wicri:level="1"><inist:fA14 i1="04"><s1>Nanjing University of Science and Technology, 200 Xiaolingwei Street</s1>
<s2>Nanjing 210094</s2>
<s3>CHN</s3>
<sZ>1 aut.</sZ>
<sZ>4 aut.</sZ>
<sZ>5 aut.</sZ>
</inist:fA14>
<country>République populaire de Chine</country>
</affiliation>
</author>
</titleStmt>
<publicationStmt><idno type="wicri:source">INIST</idno>
<idno type="inist">12-0017653</idno>
<date when="2011">2011</date>
<idno type="stanalyst">PASCAL 12-0017653 INIST</idno>
<idno type="RBID">Pascal:12-0017653</idno>
<idno type="wicri:Area/PascalFrancis/Corpus">001736</idno>
<idno type="wicri:Area/PascalFrancis/Curation">004773</idno>
</publicationStmt>
<sourceDesc><biblStruct><analytic><title xml:lang="en" level="a">Graph attribute embedding via Riemannian submersion learning</title>
<author><name sortKey="Haifeng Zhao" sort="Haifeng Zhao" uniqKey="Haifeng Zhao" last="Haifeng Zhao">HAIFENG ZHAO</name>
<affiliation wicri:level="1"><inist:fA14 i1="04"><s1>Nanjing University of Science and Technology, 200 Xiaolingwei Street</s1>
<s2>Nanjing 210094</s2>
<s3>CHN</s3>
<sZ>1 aut.</sZ>
<sZ>4 aut.</sZ>
<sZ>5 aut.</sZ>
</inist:fA14>
<country>République populaire de Chine</country>
</affiliation>
</author>
<author><name sortKey="Robles Kelly, Antonio" sort="Robles Kelly, Antonio" uniqKey="Robles Kelly A" first="Antonio" last="Robles-Kelly">Antonio Robles-Kelly</name>
<affiliation wicri:level="1"><inist:fA14 i1="01"><s1>Tower A 7 London Circuit</s1>
<s2>Canberra ACT 2601</s2>
<s3>AUS</s3>
<sZ>2 aut.</sZ>
<sZ>3 aut.</sZ>
</inist:fA14>
<country>Australie</country>
</affiliation>
<affiliation wicri:level="1"><inist:fA14 i1="02"><s1>College of Engineering and Computer Science, The Australian National University</s1>
<s2>Canberra ACT 0200</s2>
<s3>AUS</s3>
<sZ>2 aut.</sZ>
<sZ>3 aut.</sZ>
</inist:fA14>
<country>Australie</country>
</affiliation>
<affiliation wicri:level="1"><inist:fA14 i1="03"><s1>UNSW@ADFA</s1>
<s2>Canberra ACT 2600</s2>
<s3>AUS</s3>
<sZ>2 aut.</sZ>
<sZ>3 aut.</sZ>
</inist:fA14>
<country>Australie</country>
</affiliation>
</author>
<author><name sortKey="Jun Zhou" sort="Jun Zhou" uniqKey="Jun Zhou" last="Jun Zhou">JUN ZHOU</name>
<affiliation wicri:level="1"><inist:fA14 i1="01"><s1>Tower A 7 London Circuit</s1>
<s2>Canberra ACT 2601</s2>
<s3>AUS</s3>
<sZ>2 aut.</sZ>
<sZ>3 aut.</sZ>
</inist:fA14>
<country>Australie</country>
</affiliation>
<affiliation wicri:level="1"><inist:fA14 i1="02"><s1>College of Engineering and Computer Science, The Australian National University</s1>
<s2>Canberra ACT 0200</s2>
<s3>AUS</s3>
<sZ>2 aut.</sZ>
<sZ>3 aut.</sZ>
</inist:fA14>
<country>Australie</country>
</affiliation>
<affiliation wicri:level="1"><inist:fA14 i1="03"><s1>UNSW@ADFA</s1>
<s2>Canberra ACT 2600</s2>
<s3>AUS</s3>
<sZ>2 aut.</sZ>
<sZ>3 aut.</sZ>
</inist:fA14>
<country>Australie</country>
</affiliation>
</author>
<author><name sortKey="Jianfeng Lu" sort="Jianfeng Lu" uniqKey="Jianfeng Lu" last="Jianfeng Lu">JIANFENG LU</name>
<affiliation wicri:level="1"><inist:fA14 i1="04"><s1>Nanjing University of Science and Technology, 200 Xiaolingwei Street</s1>
<s2>Nanjing 210094</s2>
<s3>CHN</s3>
<sZ>1 aut.</sZ>
<sZ>4 aut.</sZ>
<sZ>5 aut.</sZ>
</inist:fA14>
<country>République populaire de Chine</country>
</affiliation>
</author>
<author><name sortKey="Yang, Jing Yu" sort="Yang, Jing Yu" uniqKey="Yang J" first="Jing-Yu" last="Yang">Jing-Yu Yang</name>
<affiliation wicri:level="1"><inist:fA14 i1="04"><s1>Nanjing University of Science and Technology, 200 Xiaolingwei Street</s1>
<s2>Nanjing 210094</s2>
<s3>CHN</s3>
<sZ>1 aut.</sZ>
<sZ>4 aut.</sZ>
<sZ>5 aut.</sZ>
</inist:fA14>
<country>République populaire de Chine</country>
</affiliation>
</author>
</analytic>
<series><title level="j" type="main">Computer vision and image understanding : (Print)</title>
<title level="j" type="abbreviated">Comput. vis. image underst. : (Print)</title>
<idno type="ISSN">1077-3142</idno>
<imprint><date when="2011">2011</date>
</imprint>
</series>
</biblStruct>
</sourceDesc>
<seriesStmt><title level="j" type="main">Computer vision and image understanding : (Print)</title>
<title level="j" type="abbreviated">Comput. vis. image underst. : (Print)</title>
<idno type="ISSN">1077-3142</idno>
</seriesStmt>
</fileDesc>
<profileDesc><textClass><keywords scheme="KwdEn" xml:lang="en"><term>Categorization</term>
<term>Classification</term>
<term>Computer vision</term>
<term>Confidence interval</term>
<term>Database</term>
<term>Density estimation</term>
<term>Density measurement</term>
<term>Graph matching</term>
<term>Image processing</term>
<term>Iterative method</term>
<term>L2 approximation</term>
<term>Metric space</term>
<term>Object recognition</term>
<term>Pattern matching</term>
<term>Posterior distribution</term>
<term>Posterior probability</term>
<term>Probability density</term>
<term>Vertex(graph)</term>
</keywords>
<keywords scheme="Pascal" xml:lang="fr"><term>Vision ordinateur</term>
<term>Concordance forme</term>
<term>Traitement image</term>
<term>Classification</term>
<term>Base de données</term>
<term>Espace métrique</term>
<term>Catégorisation</term>
<term>Estimation densité</term>
<term>Mesure densité</term>
<term>Intervalle confiance</term>
<term>Densité probabilité</term>
<term>Méthode itérative</term>
<term>Approximation L2</term>
<term>Sommet graphe</term>
<term>Probabilité a posteriori</term>
<term>Couplage graphe</term>
<term>Loi a posteriori</term>
<term>.</term>
<term>Reconnaissance objet</term>
</keywords>
<keywords scheme="Wicri" type="topic" xml:lang="fr"><term>Classification</term>
<term>Base de données</term>
</keywords>
</textClass>
</profileDesc>
</teiHeader>
<front><div type="abstract" xml:lang="en">In this paper, we tackle the problem of embedding a set of relational structures into a metric space for purposes of matching and categorisation. To this end, we view the problem from a Riemannian perspective and make use of the concepts of charts on the manifold to define the embedding as a mixture of class-specific submersions. Formulated in this manner, the mixture weights are recovered using a probability density estimation on the embedded graph node coordinates. Further, we recover these class-specific submersions making use of an iterative trust-region method so as to minimise the L2 norm between the hard limit of the graph-vertex posterior probabilities and their estimated values. The method presented here is quite general in nature and allows tasks such as matching, categorisation and retrieval. We show results on graph matching, shape categorisation and digit classification on synthetic data, the MNIST dataset and the MPEG-7 database.</div>
</front>
</TEI>
<inist><standard h6="B"><pA><fA01 i1="01" i2="1"><s0>1077-3142</s0>
</fA01>
<fA02 i1="01"><s0>CVIUF4</s0>
</fA02>
<fA03 i2="1"><s0>Comput. vis. image underst. : (Print)</s0>
</fA03>
<fA05><s2>115</s2>
</fA05>
<fA06><s2>7</s2>
</fA06>
<fA08 i1="01" i2="1" l="ENG"><s1>Graph attribute embedding via Riemannian submersion learning</s1>
</fA08>
<fA09 i1="01" i2="1" l="ENG"><s1>Special issue on Graph-Based Representations in Computer Vision</s1>
</fA09>
<fA11 i1="01" i2="1"><s1>HAIFENG ZHAO</s1>
</fA11>
<fA11 i1="02" i2="1"><s1>ROBLES-KELLY (Antonio)</s1>
</fA11>
<fA11 i1="03" i2="1"><s1>JUN ZHOU</s1>
</fA11>
<fA11 i1="04" i2="1"><s1>JIANFENG LU</s1>
</fA11>
<fA11 i1="05" i2="1"><s1>YANG (Jing-Yu)</s1>
</fA11>
<fA12 i1="01" i2="1"><s1>HANCOCK (Edwin R.)</s1>
<s9>ed.</s9>
</fA12>
<fA12 i1="02" i2="1"><s1>TORSELLO (Andrea)</s1>
<s9>ed.</s9>
</fA12>
<fA12 i1="03" i2="1"><s1>ESCOLANO (Francisco)</s1>
<s9>ed.</s9>
</fA12>
<fA12 i1="04" i2="1"><s1>BRUN (Luc)</s1>
<s9>ed.</s9>
</fA12>
<fA14 i1="01"><s1>Tower A 7 London Circuit</s1>
<s2>Canberra ACT 2601</s2>
<s3>AUS</s3>
<sZ>2 aut.</sZ>
<sZ>3 aut.</sZ>
</fA14>
<fA14 i1="02"><s1>College of Engineering and Computer Science, The Australian National University</s1>
<s2>Canberra ACT 0200</s2>
<s3>AUS</s3>
<sZ>2 aut.</sZ>
<sZ>3 aut.</sZ>
</fA14>
<fA14 i1="03"><s1>UNSW@ADFA</s1>
<s2>Canberra ACT 2600</s2>
<s3>AUS</s3>
<sZ>2 aut.</sZ>
<sZ>3 aut.</sZ>
</fA14>
<fA14 i1="04"><s1>Nanjing University of Science and Technology, 200 Xiaolingwei Street</s1>
<s2>Nanjing 210094</s2>
<s3>CHN</s3>
<sZ>1 aut.</sZ>
<sZ>4 aut.</sZ>
<sZ>5 aut.</sZ>
</fA14>
<fA15 i1="01"><s1>University of York</s1>
<s3>GBR</s3>
<sZ>1 aut.</sZ>
</fA15>
<fA15 i1="02"><s1>Università Ca' Foscari Venezia</s1>
<s3>ITA</s3>
<sZ>2 aut.</sZ>
</fA15>
<fA15 i1="03"><s1>University of Alicante</s1>
<s3>ESP</s3>
<sZ>3 aut.</sZ>
</fA15>
<fA15 i1="04"><s1>GREYC, CNRS UMR 6072, ENSICAEN</s1>
<s3>FRA</s3>
<sZ>4 aut.</sZ>
</fA15>
<fA20><s1>962-975</s1>
</fA20>
<fA21><s1>2011</s1>
</fA21>
<fA23 i1="01"><s0>ENG</s0>
</fA23>
<fA43 i1="01"><s1>INIST</s1>
<s2>15463A</s2>
<s5>354000509422430050</s5>
</fA43>
<fA44><s0>0000</s0>
<s1>© 2012 INIST-CNRS. All rights reserved.</s1>
</fA44>
<fA45><s0>66 ref.</s0>
</fA45>
<fA47 i1="01" i2="1"><s0>12-0017653</s0>
</fA47>
<fA60><s1>P</s1>
</fA60>
<fA61><s0>A</s0>
</fA61>
<fA64 i1="01" i2="1"><s0>Computer vision and image understanding : (Print)</s0>
</fA64>
<fA66 i1="01"><s0>NLD</s0>
</fA66>
<fC01 i1="01" l="ENG"><s0>In this paper, we tackle the problem of embedding a set of relational structures into a metric space for purposes of matching and categorisation. To this end, we view the problem from a Riemannian perspective and make use of the concepts of charts on the manifold to define the embedding as a mixture of class-specific submersions. Formulated in this manner, the mixture weights are recovered using a probability density estimation on the embedded graph node coordinates. Further, we recover these class-specific submersions making use of an iterative trust-region method so as to minimise the L2 norm between the hard limit of the graph-vertex posterior probabilities and their estimated values. The method presented here is quite general in nature and allows tasks such as matching, categorisation and retrieval. We show results on graph matching, shape categorisation and digit classification on synthetic data, the MNIST dataset and the MPEG-7 database.</s0>
</fC01>
<fC02 i1="01" i2="X"><s0>001D02A06</s0>
</fC02>
<fC02 i1="02" i2="X"><s0>001D02C03</s0>
</fC02>
<fC03 i1="01" i2="X" l="FRE"><s0>Vision ordinateur</s0>
<s5>01</s5>
</fC03>
<fC03 i1="01" i2="X" l="ENG"><s0>Computer vision</s0>
<s5>01</s5>
</fC03>
<fC03 i1="01" i2="X" l="SPA"><s0>Visión ordenador</s0>
<s5>01</s5>
</fC03>
<fC03 i1="02" i2="X" l="FRE"><s0>Concordance forme</s0>
<s5>06</s5>
</fC03>
<fC03 i1="02" i2="X" l="ENG"><s0>Pattern matching</s0>
<s5>06</s5>
</fC03>
<fC03 i1="03" i2="X" l="FRE"><s0>Traitement image</s0>
<s5>07</s5>
</fC03>
<fC03 i1="03" i2="X" l="ENG"><s0>Image processing</s0>
<s5>07</s5>
</fC03>
<fC03 i1="03" i2="X" l="SPA"><s0>Procesamiento imagen</s0>
<s5>07</s5>
</fC03>
<fC03 i1="04" i2="X" l="FRE"><s0>Classification</s0>
<s5>08</s5>
</fC03>
<fC03 i1="04" i2="X" l="ENG"><s0>Classification</s0>
<s5>08</s5>
</fC03>
<fC03 i1="04" i2="X" l="SPA"><s0>Clasificación</s0>
<s5>08</s5>
</fC03>
<fC03 i1="05" i2="X" l="FRE"><s0>Base de données</s0>
<s5>09</s5>
</fC03>
<fC03 i1="05" i2="X" l="ENG"><s0>Database</s0>
<s5>09</s5>
</fC03>
<fC03 i1="05" i2="X" l="SPA"><s0>Base dato</s0>
<s5>09</s5>
</fC03>
<fC03 i1="06" i2="X" l="FRE"><s0>Espace métrique</s0>
<s5>18</s5>
</fC03>
<fC03 i1="06" i2="X" l="ENG"><s0>Metric space</s0>
<s5>18</s5>
</fC03>
<fC03 i1="06" i2="X" l="SPA"><s0>Espacio métrico</s0>
<s5>18</s5>
</fC03>
<fC03 i1="07" i2="X" l="FRE"><s0>Catégorisation</s0>
<s5>19</s5>
</fC03>
<fC03 i1="07" i2="X" l="ENG"><s0>Categorization</s0>
<s5>19</s5>
</fC03>
<fC03 i1="07" i2="X" l="SPA"><s0>Categorización</s0>
<s5>19</s5>
</fC03>
<fC03 i1="08" i2="X" l="FRE"><s0>Estimation densité</s0>
<s5>20</s5>
</fC03>
<fC03 i1="08" i2="X" l="ENG"><s0>Density estimation</s0>
<s5>20</s5>
</fC03>
<fC03 i1="08" i2="X" l="SPA"><s0>Estimación densidad</s0>
<s5>20</s5>
</fC03>
<fC03 i1="09" i2="X" l="FRE"><s0>Mesure densité</s0>
<s5>21</s5>
</fC03>
<fC03 i1="09" i2="X" l="ENG"><s0>Density measurement</s0>
<s5>21</s5>
</fC03>
<fC03 i1="09" i2="X" l="SPA"><s0>Medición densidad</s0>
<s5>21</s5>
</fC03>
<fC03 i1="10" i2="X" l="FRE"><s0>Intervalle confiance</s0>
<s5>22</s5>
</fC03>
<fC03 i1="10" i2="X" l="ENG"><s0>Confidence interval</s0>
<s5>22</s5>
</fC03>
<fC03 i1="10" i2="X" l="SPA"><s0>Intervalo confianza</s0>
<s5>22</s5>
</fC03>
<fC03 i1="11" i2="X" l="FRE"><s0>Densité probabilité</s0>
<s5>23</s5>
</fC03>
<fC03 i1="11" i2="X" l="ENG"><s0>Probability density</s0>
<s5>23</s5>
</fC03>
<fC03 i1="11" i2="X" l="SPA"><s0>Densidad probabilidad</s0>
<s5>23</s5>
</fC03>
<fC03 i1="12" i2="X" l="FRE"><s0>Méthode itérative</s0>
<s5>24</s5>
</fC03>
<fC03 i1="12" i2="X" l="ENG"><s0>Iterative method</s0>
<s5>24</s5>
</fC03>
<fC03 i1="12" i2="X" l="SPA"><s0>Método iterativo</s0>
<s5>24</s5>
</fC03>
<fC03 i1="13" i2="X" l="FRE"><s0>Approximation L2</s0>
<s5>25</s5>
</fC03>
<fC03 i1="13" i2="X" l="ENG"><s0>L2 approximation</s0>
<s5>25</s5>
</fC03>
<fC03 i1="13" i2="X" l="SPA"><s0>Aproximación L2</s0>
<s5>25</s5>
</fC03>
<fC03 i1="14" i2="X" l="FRE"><s0>Sommet graphe</s0>
<s5>26</s5>
</fC03>
<fC03 i1="14" i2="X" l="ENG"><s0>Vertex(graph)</s0>
<s5>26</s5>
</fC03>
<fC03 i1="14" i2="X" l="SPA"><s0>Vértice grafo</s0>
<s5>26</s5>
</fC03>
<fC03 i1="15" i2="X" l="FRE"><s0>Probabilité a posteriori</s0>
<s5>27</s5>
</fC03>
<fC03 i1="15" i2="X" l="ENG"><s0>Posterior probability</s0>
<s5>27</s5>
</fC03>
<fC03 i1="15" i2="X" l="SPA"><s0>Probabilidad a posteriori</s0>
<s5>27</s5>
</fC03>
<fC03 i1="16" i2="X" l="FRE"><s0>Couplage graphe</s0>
<s5>28</s5>
</fC03>
<fC03 i1="16" i2="X" l="ENG"><s0>Graph matching</s0>
<s5>28</s5>
</fC03>
<fC03 i1="16" i2="X" l="SPA"><s0>Acoplamiento grafo</s0>
<s5>28</s5>
</fC03>
<fC03 i1="17" i2="X" l="FRE"><s0>Loi a posteriori</s0>
<s5>41</s5>
</fC03>
<fC03 i1="17" i2="X" l="ENG"><s0>Posterior distribution</s0>
<s5>41</s5>
</fC03>
<fC03 i1="17" i2="X" l="SPA"><s0>Ley a posteriori</s0>
<s5>41</s5>
</fC03>
<fC03 i1="18" i2="X" l="FRE"><s0>.</s0>
<s4>INC</s4>
<s5>82</s5>
</fC03>
<fC03 i1="19" i2="X" l="FRE"><s0>Reconnaissance objet</s0>
<s4>CD</s4>
<s5>96</s5>
</fC03>
<fC03 i1="19" i2="X" l="ENG"><s0>Object recognition</s0>
<s4>CD</s4>
<s5>96</s5>
</fC03>
<fC03 i1="19" i2="X" l="SPA"><s0>Reconocimiento de objetos</s0>
<s4>CD</s4>
<s5>96</s5>
</fC03>
<fN21><s1>002</s1>
</fN21>
<fN44 i1="01"><s1>OTO</s1>
</fN44>
<fN82><s1>OTO</s1>
</fN82>
</pA>
</standard>
</inist>
</record>

Pour manipuler ce document sous Unix (Dilib)

EXPLOR_STEP=$WICRI_ROOT/Wicri/Asie/explor/AustralieFrV1/Data/PascalFrancis/Curation

HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 004773 | SxmlIndent | more

HfdSelect -h $EXPLOR_AREA/Data/PascalFrancis/Curation/biblio.hfd -nk 004773 | SxmlIndent | more

Pour mettre un lien sur cette page dans le réseau Wicri

{{Explor lien
   |wiki=    Wicri/Asie
   |area=    AustralieFrV1
   |flux=    PascalFrancis
   |étape=   Curation
   |type=    RBID
   |clé=     Pascal:12-0017653
   |texte=   Graph attribute embedding via Riemannian submersion learning
}}

This area was generated with Dilib version V0.6.33.
Data generation: Tue Dec 5 10:43:12 2017. Site generation: Tue Mar 5 14:07:20 2024

	Serveur d'exploration sur les relations entre la France et l'Australie
	Attention, ce site est en cours de développement ! Attention, site généré par des moyens informatiques à partir de corpus bruts. Les informations ne sont donc pas validées.

Serveur d'exploration sur les relations entre la France et l'Australie

Graph attribute embedding via Riemannian submersion learning

Graph attribute embedding via Riemannian submersion learning

Source :

Descripteurs français

English descriptors

Abstract

Links toward previous steps (curation, corpus...)

Links to Exploration step

Le document en format XML

Pour manipuler ce document sous Unix (Dilib)

Pour mettre un lien sur cette page dans le réseau Wicri