Evidence-driven image interpretation by combining implicit and explicit knowledge in a Bayesian network.
Identifieur interne : 000734 ( Ncbi/Merge ); précédent : 000733; suivant : 000735Evidence-driven image interpretation by combining implicit and explicit knowledge in a Bayesian network.
Auteurs : Spiros Nikolopoulos [Grèce] ; Georgios Th Papadopoulos ; Ioannis Kompatsiaris ; Ioannis PatrasSource :
- IEEE transactions on systems, man, and cybernetics. Part B, Cybernetics : a publication of the IEEE Systems, Man, and Cybernetics Society [ 1941-0492 ] ; 2011.
Descripteurs français
- KwdFr :
- MESH :
English descriptors
- KwdEn :
- MESH :
Abstract
Computer vision techniques have made considerable progress in recognizing object categories by learning models that normally rely on a set of discriminative features. However, in contrast to human perception that makes extensive use of logic-based rules, these models fail to benefit from knowledge that is explicitly provided. In this paper, we propose a framework that can perform knowledge-assisted analysis of visual content. We use ontologies to model the domain knowledge and a set of conditional probabilities to model the application context. Then, a Bayesian network is used for integrating statistical and explicit knowledge and performing hypothesis testing using evidence-driven probabilistic inference. In addition, we propose the use of a focus-of-attention (FoA) mechanism that is based on the mutual information between concepts. This mechanism selects the most prominent hypotheses to be verified/tested by the BN, hence removing the need to exhaustively test all possible combinations of the hypotheses set. We experimentally evaluate our framework using content from three domains and for the following three tasks: 1) image categorization; 2) localized region labeling; and 3) weak annotation of video shot keyframes. The results obtained demonstrate the improvement in performance compared to a set of baseline concept classifiers that are not aware of any context or domain knowledge. Finally, we also demonstrate the ability of the proposed FoA mechanism to significantly reduce the computational cost of visual inference while obtaining results comparable to the exhaustive case.
DOI: 10.1109/TSMCB.2011.2147781
PubMed: 21642042
Links toward previous steps (curation, corpus...)
- to stream PubMed, to step Corpus: 000324
- to stream PubMed, to step Curation: 000324
- to stream PubMed, to step Checkpoint: 000324
Links to Exploration step
pubmed:21642042Le document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en">Evidence-driven image interpretation by combining implicit and explicit knowledge in a Bayesian network.</title>
<author><name sortKey="Nikolopoulos, Spiros" sort="Nikolopoulos, Spiros" uniqKey="Nikolopoulos S" first="Spiros" last="Nikolopoulos">Spiros Nikolopoulos</name>
<affiliation wicri:level="1"><nlm:affiliation>Centre for Research and Technology Hellas/Informatics and Telematics Institute (CERTH/ITI), Thessaloniki, Greece. nikolopo@iti.gr</nlm:affiliation>
<country xml:lang="fr">Grèce</country>
<wicri:regionArea>Centre for Research and Technology Hellas/Informatics and Telematics Institute (CERTH/ITI), Thessaloniki</wicri:regionArea>
<wicri:noRegion>Thessaloniki</wicri:noRegion>
</affiliation>
</author>
<author><name sortKey="Papadopoulos, Georgios Th" sort="Papadopoulos, Georgios Th" uniqKey="Papadopoulos G" first="Georgios Th" last="Papadopoulos">Georgios Th Papadopoulos</name>
</author>
<author><name sortKey="Kompatsiaris, Ioannis" sort="Kompatsiaris, Ioannis" uniqKey="Kompatsiaris I" first="Ioannis" last="Kompatsiaris">Ioannis Kompatsiaris</name>
</author>
<author><name sortKey="Patras, Ioannis" sort="Patras, Ioannis" uniqKey="Patras I" first="Ioannis" last="Patras">Ioannis Patras</name>
</author>
</titleStmt>
<publicationStmt><idno type="wicri:source">PubMed</idno>
<date when="2011">2011</date>
<idno type="doi">10.1109/TSMCB.2011.2147781</idno>
<idno type="RBID">pubmed:21642042</idno>
<idno type="pmid">21642042</idno>
<idno type="wicri:Area/PubMed/Corpus">000324</idno>
<idno type="wicri:explorRef" wicri:stream="PubMed" wicri:step="Corpus" wicri:corpus="PubMed">000324</idno>
<idno type="wicri:Area/PubMed/Curation">000324</idno>
<idno type="wicri:explorRef" wicri:stream="PubMed" wicri:step="Curation">000324</idno>
<idno type="wicri:Area/PubMed/Checkpoint">000324</idno>
<idno type="wicri:explorRef" wicri:stream="Checkpoint" wicri:step="PubMed">000324</idno>
<idno type="wicri:Area/Ncbi/Merge">000734</idno>
</publicationStmt>
<sourceDesc><biblStruct><analytic><title xml:lang="en">Evidence-driven image interpretation by combining implicit and explicit knowledge in a Bayesian network.</title>
<author><name sortKey="Nikolopoulos, Spiros" sort="Nikolopoulos, Spiros" uniqKey="Nikolopoulos S" first="Spiros" last="Nikolopoulos">Spiros Nikolopoulos</name>
<affiliation wicri:level="1"><nlm:affiliation>Centre for Research and Technology Hellas/Informatics and Telematics Institute (CERTH/ITI), Thessaloniki, Greece. nikolopo@iti.gr</nlm:affiliation>
<country xml:lang="fr">Grèce</country>
<wicri:regionArea>Centre for Research and Technology Hellas/Informatics and Telematics Institute (CERTH/ITI), Thessaloniki</wicri:regionArea>
<wicri:noRegion>Thessaloniki</wicri:noRegion>
</affiliation>
</author>
<author><name sortKey="Papadopoulos, Georgios Th" sort="Papadopoulos, Georgios Th" uniqKey="Papadopoulos G" first="Georgios Th" last="Papadopoulos">Georgios Th Papadopoulos</name>
</author>
<author><name sortKey="Kompatsiaris, Ioannis" sort="Kompatsiaris, Ioannis" uniqKey="Kompatsiaris I" first="Ioannis" last="Kompatsiaris">Ioannis Kompatsiaris</name>
</author>
<author><name sortKey="Patras, Ioannis" sort="Patras, Ioannis" uniqKey="Patras I" first="Ioannis" last="Patras">Ioannis Patras</name>
</author>
</analytic>
<series><title level="j">IEEE transactions on systems, man, and cybernetics. Part B, Cybernetics : a publication of the IEEE Systems, Man, and Cybernetics Society</title>
<idno type="eISSN">1941-0492</idno>
<imprint><date when="2011" type="published">2011</date>
</imprint>
</series>
</biblStruct>
</sourceDesc>
</fileDesc>
<profileDesc><textClass><keywords scheme="KwdEn" xml:lang="en"><term>Algorithms</term>
<term>Attention</term>
<term>Bayes Theorem</term>
<term>Cybernetics</term>
<term>Image Processing, Computer-Assisted (methods)</term>
<term>Neural Networks (Computer)</term>
<term>Probability</term>
<term>Visual Perception</term>
</keywords>
<keywords scheme="KwdFr" xml:lang="fr"><term>Algorithmes</term>
<term>Attention</term>
<term>Cybernétique</term>
<term>Perception visuelle</term>
<term>Probabilité</term>
<term>Réseaux neuronaux (ordinateur)</term>
<term>Théorème de Bayes</term>
<term>Traitement d'image par ordinateur ()</term>
</keywords>
<keywords scheme="MESH" qualifier="methods" xml:lang="en"><term>Image Processing, Computer-Assisted</term>
</keywords>
<keywords scheme="MESH" xml:lang="en"><term>Algorithms</term>
<term>Attention</term>
<term>Bayes Theorem</term>
<term>Cybernetics</term>
<term>Neural Networks (Computer)</term>
<term>Probability</term>
<term>Visual Perception</term>
</keywords>
<keywords scheme="MESH" xml:lang="fr"><term>Algorithmes</term>
<term>Attention</term>
<term>Cybernétique</term>
<term>Perception visuelle</term>
<term>Probabilité</term>
<term>Réseaux neuronaux (ordinateur)</term>
<term>Théorème de Bayes</term>
<term>Traitement d'image par ordinateur</term>
</keywords>
</textClass>
</profileDesc>
</teiHeader>
<front><div type="abstract" xml:lang="en">Computer vision techniques have made considerable progress in recognizing object categories by learning models that normally rely on a set of discriminative features. However, in contrast to human perception that makes extensive use of logic-based rules, these models fail to benefit from knowledge that is explicitly provided. In this paper, we propose a framework that can perform knowledge-assisted analysis of visual content. We use ontologies to model the domain knowledge and a set of conditional probabilities to model the application context. Then, a Bayesian network is used for integrating statistical and explicit knowledge and performing hypothesis testing using evidence-driven probabilistic inference. In addition, we propose the use of a focus-of-attention (FoA) mechanism that is based on the mutual information between concepts. This mechanism selects the most prominent hypotheses to be verified/tested by the BN, hence removing the need to exhaustively test all possible combinations of the hypotheses set. We experimentally evaluate our framework using content from three domains and for the following three tasks: 1) image categorization; 2) localized region labeling; and 3) weak annotation of video shot keyframes. The results obtained demonstrate the improvement in performance compared to a set of baseline concept classifiers that are not aware of any context or domain knowledge. Finally, we also demonstrate the ability of the proposed FoA mechanism to significantly reduce the computational cost of visual inference while obtaining results comparable to the exhaustive case.</div>
</front>
</TEI>
<pubmed><MedlineCitation Owner="NLM" Status="MEDLINE"><PMID Version="1">21642042</PMID>
<DateCreated><Year>2011</Year>
<Month>09</Month>
<Day>19</Day>
</DateCreated>
<DateCompleted><Year>2012</Year>
<Month>01</Month>
<Day>18</Day>
</DateCompleted>
<Article PubModel="Print-Electronic"><Journal><ISSN IssnType="Electronic">1941-0492</ISSN>
<JournalIssue CitedMedium="Internet"><Volume>41</Volume>
<Issue>5</Issue>
<PubDate><Year>2011</Year>
<Month>Oct</Month>
</PubDate>
</JournalIssue>
<Title>IEEE transactions on systems, man, and cybernetics. Part B, Cybernetics : a publication of the IEEE Systems, Man, and Cybernetics Society</Title>
<ISOAbbreviation>IEEE Trans Syst Man Cybern B Cybern</ISOAbbreviation>
</Journal>
<ArticleTitle>Evidence-driven image interpretation by combining implicit and explicit knowledge in a Bayesian network.</ArticleTitle>
<Pagination><MedlinePgn>1366-81</MedlinePgn>
</Pagination>
<ELocationID EIdType="doi" ValidYN="Y">10.1109/TSMCB.2011.2147781</ELocationID>
<Abstract><AbstractText>Computer vision techniques have made considerable progress in recognizing object categories by learning models that normally rely on a set of discriminative features. However, in contrast to human perception that makes extensive use of logic-based rules, these models fail to benefit from knowledge that is explicitly provided. In this paper, we propose a framework that can perform knowledge-assisted analysis of visual content. We use ontologies to model the domain knowledge and a set of conditional probabilities to model the application context. Then, a Bayesian network is used for integrating statistical and explicit knowledge and performing hypothesis testing using evidence-driven probabilistic inference. In addition, we propose the use of a focus-of-attention (FoA) mechanism that is based on the mutual information between concepts. This mechanism selects the most prominent hypotheses to be verified/tested by the BN, hence removing the need to exhaustively test all possible combinations of the hypotheses set. We experimentally evaluate our framework using content from three domains and for the following three tasks: 1) image categorization; 2) localized region labeling; and 3) weak annotation of video shot keyframes. The results obtained demonstrate the improvement in performance compared to a set of baseline concept classifiers that are not aware of any context or domain knowledge. Finally, we also demonstrate the ability of the proposed FoA mechanism to significantly reduce the computational cost of visual inference while obtaining results comparable to the exhaustive case.</AbstractText>
</Abstract>
<AuthorList CompleteYN="Y"><Author ValidYN="Y"><LastName>Nikolopoulos</LastName>
<ForeName>Spiros</ForeName>
<Initials>S</Initials>
<AffiliationInfo><Affiliation>Centre for Research and Technology Hellas/Informatics and Telematics Institute (CERTH/ITI), Thessaloniki, Greece. nikolopo@iti.gr</Affiliation>
</AffiliationInfo>
</Author>
<Author ValidYN="Y"><LastName>Papadopoulos</LastName>
<ForeName>Georgios Th</ForeName>
<Initials>GT</Initials>
</Author>
<Author ValidYN="Y"><LastName>Kompatsiaris</LastName>
<ForeName>Ioannis</ForeName>
<Initials>I</Initials>
</Author>
<Author ValidYN="Y"><LastName>Patras</LastName>
<ForeName>Ioannis</ForeName>
<Initials>I</Initials>
</Author>
</AuthorList>
<Language>eng</Language>
<PublicationTypeList><PublicationType UI="D016428">Journal Article</PublicationType>
<PublicationType UI="D013485">Research Support, Non-U.S. Gov't</PublicationType>
</PublicationTypeList>
<ArticleDate DateType="Electronic"><Year>2011</Year>
<Month>06</Month>
<Day>02</Day>
</ArticleDate>
</Article>
<MedlineJournalInfo><Country>United States</Country>
<MedlineTA>IEEE Trans Syst Man Cybern B Cybern</MedlineTA>
<NlmUniqueID>9890044</NlmUniqueID>
<ISSNLinking>1083-4419</ISSNLinking>
</MedlineJournalInfo>
<CitationSubset>IM</CitationSubset>
<MeshHeadingList><MeshHeading><DescriptorName MajorTopicYN="N" UI="D000465">Algorithms</DescriptorName>
</MeshHeading>
<MeshHeading><DescriptorName MajorTopicYN="N" UI="D001288">Attention</DescriptorName>
</MeshHeading>
<MeshHeading><DescriptorName MajorTopicYN="Y" UI="D001499">Bayes Theorem</DescriptorName>
</MeshHeading>
<MeshHeading><DescriptorName MajorTopicYN="Y" UI="D003491">Cybernetics</DescriptorName>
</MeshHeading>
<MeshHeading><DescriptorName MajorTopicYN="N" UI="D007091">Image Processing, Computer-Assisted</DescriptorName>
<QualifierName MajorTopicYN="Y" UI="Q000379">methods</QualifierName>
</MeshHeading>
<MeshHeading><DescriptorName MajorTopicYN="Y" UI="D016571">Neural Networks (Computer)</DescriptorName>
</MeshHeading>
<MeshHeading><DescriptorName MajorTopicYN="N" UI="D011336">Probability</DescriptorName>
</MeshHeading>
<MeshHeading><DescriptorName MajorTopicYN="N" UI="D014796">Visual Perception</DescriptorName>
</MeshHeading>
</MeshHeadingList>
</MedlineCitation>
<PubmedData><History><PubMedPubDate PubStatus="aheadofprint"><Year>2011</Year>
<Month>6</Month>
<Day>02</Day>
</PubMedPubDate>
<PubMedPubDate PubStatus="entrez"><Year>2011</Year>
<Month>6</Month>
<Day>7</Day>
<Hour>6</Hour>
<Minute>0</Minute>
</PubMedPubDate>
<PubMedPubDate PubStatus="pubmed"><Year>2011</Year>
<Month>6</Month>
<Day>7</Day>
<Hour>6</Hour>
<Minute>0</Minute>
</PubMedPubDate>
<PubMedPubDate PubStatus="medline"><Year>2012</Year>
<Month>1</Month>
<Day>19</Day>
<Hour>6</Hour>
<Minute>0</Minute>
</PubMedPubDate>
</History>
<PublicationStatus>ppublish</PublicationStatus>
<ArticleIdList><ArticleId IdType="doi">10.1109/TSMCB.2011.2147781</ArticleId>
<ArticleId IdType="pubmed">21642042</ArticleId>
</ArticleIdList>
</PubmedData>
</pubmed>
<affiliations><list><country><li>Grèce</li>
</country>
</list>
<tree><noCountry><name sortKey="Kompatsiaris, Ioannis" sort="Kompatsiaris, Ioannis" uniqKey="Kompatsiaris I" first="Ioannis" last="Kompatsiaris">Ioannis Kompatsiaris</name>
<name sortKey="Papadopoulos, Georgios Th" sort="Papadopoulos, Georgios Th" uniqKey="Papadopoulos G" first="Georgios Th" last="Papadopoulos">Georgios Th Papadopoulos</name>
<name sortKey="Patras, Ioannis" sort="Patras, Ioannis" uniqKey="Patras I" first="Ioannis" last="Patras">Ioannis Patras</name>
</noCountry>
<country name="Grèce"><noRegion><name sortKey="Nikolopoulos, Spiros" sort="Nikolopoulos, Spiros" uniqKey="Nikolopoulos S" first="Spiros" last="Nikolopoulos">Spiros Nikolopoulos</name>
</noRegion>
</country>
</tree>
</affiliations>
</record>
Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Ticri/CIDE/explor/TelematiV1/Data/Ncbi/Merge
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 000734 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Ncbi/Merge/biblio.hfd -nk 000734 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien |wiki= Ticri/CIDE |area= TelematiV1 |flux= Ncbi |étape= Merge |type= RBID |clé= pubmed:21642042 |texte= Evidence-driven image interpretation by combining implicit and explicit knowledge in a Bayesian network. }}
Pour générer des pages wiki
HfdIndexSelect -h $EXPLOR_AREA/Data/Ncbi/Merge/RBID.i -Sk "pubmed:21642042" \ | HfdSelect -Kh $EXPLOR_AREA/Data/Ncbi/Merge/biblio.hfd \ | NlmPubMed2Wicri -a TelematiV1
![]() | This area was generated with Dilib version V0.6.31. | ![]() |