Context-Based Support Vector Machines for Interconnected Image Annotation
Identifieur interne : 006465 ( Main/Exploration ); précédent : 006464; suivant : 006466Context-Based Support Vector Machines for Interconnected Image Annotation
Auteurs : Hichem Sahbi [France] ; Xi Li [France, Australie, République populaire de Chine]Source :
- Lecture Notes in Computer Science [ 0302-9743 ] ; 2011.
Descripteurs français
- Wicri :
- topic : Base de données, Traduction automatique, Multimédia.
English descriptors
- KwdEn :
- Annotation, Annotation methods, Annotation performance, Annotation performances, Annotation results, Automatic image annotation, Baseline kernel, Baseline kernels, Causal strength, Concept indices, Context criterion, Contextdependent kernel comparison, Contextual information, Database, Datasets, Energy function, Flickr, Flickr distance, Google similarity distance, Gram matrix, High value, High values, Image annotation, Image database, Image retrieval, Images share, Kernel, Kernel design, Level features, Machine translation, Matrix, Minimization problem, Multimedia, Nuswide, Nuswide dataset, Nuswide datasets, Original database, Proc, Relative gain, Resp, Sahbi, Same order, Similarity measure, Small value, Social networks, Spatial pyramid levels, Standard deviations, Support vector machines, Svms, Textual feature, Textual features, Visual content, Visual feature, Visual features.
- Teeft :
- Annotation, Annotation methods, Annotation performance, Annotation performances, Annotation results, Automatic image annotation, Baseline kernel, Baseline kernels, Causal strength, Concept indices, Context criterion, Contextdependent kernel comparison, Contextual information, Database, Datasets, Energy function, Flickr, Flickr distance, Google similarity distance, Gram matrix, High value, High values, Image annotation, Image database, Image retrieval, Images share, Kernel, Kernel design, Level features, Machine translation, Matrix, Minimization problem, Multimedia, Nuswide, Nuswide dataset, Nuswide datasets, Original database, Proc, Relative gain, Resp, Sahbi, Same order, Similarity measure, Small value, Social networks, Spatial pyramid levels, Standard deviations, Support vector machines, Svms, Textual feature, Textual features, Visual content, Visual feature, Visual features.
Abstract
Abstract: We introduce in this paper a novel image annotation approach based on support vector machines (SVMs) and a new class of kernels referred to as context-dependent. The method goes beyond the naive use of the intrinsic low level features (such as color, texture, shape, etc.) and context-free kernels, in order to design a kernel function applicable to interconnected databases such as social networks. The main contribution of our method includes (i) a variational approach which helps designing this function using both intrinsic features and the underlying contextual information resulting from different links and (ii) the proof of convergence of our kernel to a positive definite fixed-point, usable for SVM training and other kernel methods. When plugged in SVMs, our context-dependent kernel consistently improves the performance of image annotation, compared to context-free kernels, on hundreds of thousands of Flickr images.
Url:
DOI: 10.1007/978-3-642-19315-6_17
Affiliations:
Links toward previous steps (curation, corpus...)
- to stream Istex, to step Corpus: 001979
- to stream Istex, to step Curation: 001979
- to stream Istex, to step Checkpoint: 000885
- to stream Main, to step Merge: 006842
- to stream Main, to step Curation: 006465
Le document en format XML
<record><TEI wicri:istexFullTextTei="biblStruct"><teiHeader><fileDesc><titleStmt><title xml:lang="en">Context-Based Support Vector Machines for Interconnected Image Annotation</title>
<author><name sortKey="Sahbi, Hichem" sort="Sahbi, Hichem" uniqKey="Sahbi H" first="Hichem" last="Sahbi">Hichem Sahbi</name>
</author>
<author><name sortKey="Li, Xi" sort="Li, Xi" uniqKey="Li X" first="Xi" last="Li">Xi Li</name>
</author>
</titleStmt>
<publicationStmt><idno type="wicri:source">ISTEX</idno>
<idno type="RBID">ISTEX:88D346215897958587666610EAD46396BBF0CD3B</idno>
<date when="2011" year="2011">2011</date>
<idno type="doi">10.1007/978-3-642-19315-6_17</idno>
<idno type="url">https://api.istex.fr/document/88D346215897958587666610EAD46396BBF0CD3B/fulltext/pdf</idno>
<idno type="wicri:Area/Istex/Corpus">001979</idno>
<idno type="wicri:explorRef" wicri:stream="Istex" wicri:step="Corpus" wicri:corpus="ISTEX">001979</idno>
<idno type="wicri:Area/Istex/Curation">001979</idno>
<idno type="wicri:Area/Istex/Checkpoint">000885</idno>
<idno type="wicri:explorRef" wicri:stream="Istex" wicri:step="Checkpoint">000885</idno>
<idno type="wicri:doubleKey">0302-9743:2011:Sahbi H:context:based:support</idno>
<idno type="wicri:Area/Main/Merge">006842</idno>
<idno type="wicri:Area/Main/Curation">006465</idno>
<idno type="wicri:Area/Main/Exploration">006465</idno>
</publicationStmt>
<sourceDesc><biblStruct><analytic><title level="a" type="main" xml:lang="en">Context-Based Support Vector Machines for Interconnected Image Annotation</title>
<author><name sortKey="Sahbi, Hichem" sort="Sahbi, Hichem" uniqKey="Sahbi H" first="Hichem" last="Sahbi">Hichem Sahbi</name>
<affiliation wicri:level="3"><country xml:lang="fr">France</country>
<wicri:regionArea>CNRS Telecom ParisTech, Paris</wicri:regionArea>
<placeName><region type="region">Île-de-France</region>
<region type="old region">Île-de-France</region>
<settlement type="city">Paris</settlement>
</placeName>
</affiliation>
<affiliation wicri:level="1"><country wicri:rule="url">France</country>
</affiliation>
</author>
<author><name sortKey="Li, Xi" sort="Li, Xi" uniqKey="Li X" first="Xi" last="Li">Xi Li</name>
<affiliation wicri:level="3"><country xml:lang="fr">France</country>
<wicri:regionArea>CNRS Telecom ParisTech, Paris</wicri:regionArea>
<placeName><region type="region">Île-de-France</region>
<region type="old region">Île-de-France</region>
<settlement type="city">Paris</settlement>
</placeName>
</affiliation>
<affiliation wicri:level="1"><country xml:lang="fr">Australie</country>
<wicri:regionArea>School of Computer Science, The University of Adelaide</wicri:regionArea>
<wicri:noRegion>The University of Adelaide</wicri:noRegion>
</affiliation>
<affiliation wicri:level="3"><country xml:lang="fr">République populaire de Chine</country>
<wicri:regionArea>NLPR, CASIA, Beijing</wicri:regionArea>
<placeName><settlement type="city">Pékin</settlement>
</placeName>
</affiliation>
<affiliation></affiliation>
</author>
</analytic>
<monogr></monogr>
<series><title level="s">Lecture Notes in Computer Science</title>
<imprint><date>2011</date>
</imprint>
<idno type="ISSN">0302-9743</idno>
<idno type="eISSN">1611-3349</idno>
<idno type="ISSN">0302-9743</idno>
</series>
</biblStruct>
</sourceDesc>
<seriesStmt><idno type="ISSN">0302-9743</idno>
</seriesStmt>
</fileDesc>
<profileDesc><textClass><keywords scheme="KwdEn" xml:lang="en"><term>Annotation</term>
<term>Annotation methods</term>
<term>Annotation performance</term>
<term>Annotation performances</term>
<term>Annotation results</term>
<term>Automatic image annotation</term>
<term>Baseline kernel</term>
<term>Baseline kernels</term>
<term>Causal strength</term>
<term>Concept indices</term>
<term>Context criterion</term>
<term>Contextdependent kernel comparison</term>
<term>Contextual information</term>
<term>Database</term>
<term>Datasets</term>
<term>Energy function</term>
<term>Flickr</term>
<term>Flickr distance</term>
<term>Google similarity distance</term>
<term>Gram matrix</term>
<term>High value</term>
<term>High values</term>
<term>Image annotation</term>
<term>Image database</term>
<term>Image retrieval</term>
<term>Images share</term>
<term>Kernel</term>
<term>Kernel design</term>
<term>Level features</term>
<term>Machine translation</term>
<term>Matrix</term>
<term>Minimization problem</term>
<term>Multimedia</term>
<term>Nuswide</term>
<term>Nuswide dataset</term>
<term>Nuswide datasets</term>
<term>Original database</term>
<term>Proc</term>
<term>Relative gain</term>
<term>Resp</term>
<term>Sahbi</term>
<term>Same order</term>
<term>Similarity measure</term>
<term>Small value</term>
<term>Social networks</term>
<term>Spatial pyramid levels</term>
<term>Standard deviations</term>
<term>Support vector machines</term>
<term>Svms</term>
<term>Textual feature</term>
<term>Textual features</term>
<term>Visual content</term>
<term>Visual feature</term>
<term>Visual features</term>
</keywords>
<keywords scheme="Teeft" xml:lang="en"><term>Annotation</term>
<term>Annotation methods</term>
<term>Annotation performance</term>
<term>Annotation performances</term>
<term>Annotation results</term>
<term>Automatic image annotation</term>
<term>Baseline kernel</term>
<term>Baseline kernels</term>
<term>Causal strength</term>
<term>Concept indices</term>
<term>Context criterion</term>
<term>Contextdependent kernel comparison</term>
<term>Contextual information</term>
<term>Database</term>
<term>Datasets</term>
<term>Energy function</term>
<term>Flickr</term>
<term>Flickr distance</term>
<term>Google similarity distance</term>
<term>Gram matrix</term>
<term>High value</term>
<term>High values</term>
<term>Image annotation</term>
<term>Image database</term>
<term>Image retrieval</term>
<term>Images share</term>
<term>Kernel</term>
<term>Kernel design</term>
<term>Level features</term>
<term>Machine translation</term>
<term>Matrix</term>
<term>Minimization problem</term>
<term>Multimedia</term>
<term>Nuswide</term>
<term>Nuswide dataset</term>
<term>Nuswide datasets</term>
<term>Original database</term>
<term>Proc</term>
<term>Relative gain</term>
<term>Resp</term>
<term>Sahbi</term>
<term>Same order</term>
<term>Similarity measure</term>
<term>Small value</term>
<term>Social networks</term>
<term>Spatial pyramid levels</term>
<term>Standard deviations</term>
<term>Support vector machines</term>
<term>Svms</term>
<term>Textual feature</term>
<term>Textual features</term>
<term>Visual content</term>
<term>Visual feature</term>
<term>Visual features</term>
</keywords>
<keywords scheme="Wicri" type="topic" xml:lang="fr"><term>Base de données</term>
<term>Traduction automatique</term>
<term>Multimédia</term>
</keywords>
</textClass>
<langUsage><language ident="en">en</language>
</langUsage>
</profileDesc>
</teiHeader>
<front><div type="abstract" xml:lang="en">Abstract: We introduce in this paper a novel image annotation approach based on support vector machines (SVMs) and a new class of kernels referred to as context-dependent. The method goes beyond the naive use of the intrinsic low level features (such as color, texture, shape, etc.) and context-free kernels, in order to design a kernel function applicable to interconnected databases such as social networks. The main contribution of our method includes (i) a variational approach which helps designing this function using both intrinsic features and the underlying contextual information resulting from different links and (ii) the proof of convergence of our kernel to a positive definite fixed-point, usable for SVM training and other kernel methods. When plugged in SVMs, our context-dependent kernel consistently improves the performance of image annotation, compared to context-free kernels, on hundreds of thousands of Flickr images.</div>
</front>
</TEI>
<affiliations><list><country><li>Australie</li>
<li>France</li>
<li>République populaire de Chine</li>
</country>
<region><li>Île-de-France</li>
</region>
<settlement><li>Paris</li>
<li>Pékin</li>
</settlement>
</list>
<tree><country name="France"><region name="Île-de-France"><name sortKey="Sahbi, Hichem" sort="Sahbi, Hichem" uniqKey="Sahbi H" first="Hichem" last="Sahbi">Hichem Sahbi</name>
</region>
<name sortKey="Li, Xi" sort="Li, Xi" uniqKey="Li X" first="Xi" last="Li">Xi Li</name>
<name sortKey="Sahbi, Hichem" sort="Sahbi, Hichem" uniqKey="Sahbi H" first="Hichem" last="Sahbi">Hichem Sahbi</name>
</country>
<country name="Australie"><noRegion><name sortKey="Li, Xi" sort="Li, Xi" uniqKey="Li X" first="Xi" last="Li">Xi Li</name>
</noRegion>
</country>
<country name="République populaire de Chine"><noRegion><name sortKey="Li, Xi" sort="Li, Xi" uniqKey="Li X" first="Xi" last="Li">Xi Li</name>
</noRegion>
</country>
</tree>
</affiliations>
</record>
Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Wicri/Asie/explor/AustralieFrV1/Data/Main/Exploration
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 006465 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Main/Exploration/biblio.hfd -nk 006465 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien |wiki= Wicri/Asie |area= AustralieFrV1 |flux= Main |étape= Exploration |type= RBID |clé= ISTEX:88D346215897958587666610EAD46396BBF0CD3B |texte= Context-Based Support Vector Machines for Interconnected Image Annotation }}
This area was generated with Dilib version V0.6.33. |