Show, Match and Segment: Joint Weakly Supervised Learning of Semantic Matching and Object Co-segmentation.
Identifieur interne : 000077 ( Main/Exploration ); précédent : 000076; suivant : 000078Show, Match and Segment: Joint Weakly Supervised Learning of Semantic Matching and Object Co-segmentation.
Auteurs : Yun-Chun Chen ; Yen-Yu Lin ; Ming-Hsuan Yang ; Jia-Bin HuangSource :
- IEEE transactions on pattern analysis and machine intelligence [ 1939-3539 ] ; 2020.
Abstract
We present an approach for jointly matching and segmenting object instances of the same category within a collection of images. In contrast to existing algorithms that tackle the tasks of semantic matching and object co-segmentation in isolation, our method exploits the complementary nature of the two tasks. The key insights of our method are two-fold. First, the estimated dense correspondence fields from semantic matching provide supervision for object co-segmentation by enforcing consistency between the predicted masks from a pair of images. Second, the predicted object masks from object co-segmentation, in turn, allow us to reduce the adverse effects due to background clutters for improving semantic matching. Our model is end-to-end trainable and does not require supervision from manually annotated correspondences and object masks. We validate the efficacy of our approach on five benchmark datasets: TSS, Internet, PF-PASCAL, PF-WILLOW, and SPair-71k, and show that our algorithm performs favorably against the state-of-the-art methods on both semantic matching and object co-segmentation tasks.
DOI: 10.1109/TPAMI.2020.2985395
PubMed: 32275584
Affiliations:
Links toward previous steps (curation, corpus...)
Le document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en">Show, Match and Segment: Joint Weakly Supervised Learning of Semantic Matching and Object Co-segmentation.</title><author><name sortKey="Chen, Yun Chun" sort="Chen, Yun Chun" uniqKey="Chen Y" first="Yun-Chun" last="Chen">Yun-Chun Chen</name></author><author><name sortKey="Lin, Yen Yu" sort="Lin, Yen Yu" uniqKey="Lin Y" first="Yen-Yu" last="Lin">Yen-Yu Lin</name></author><author><name sortKey="Yang, Ming Hsuan" sort="Yang, Ming Hsuan" uniqKey="Yang M" first="Ming-Hsuan" last="Yang">Ming-Hsuan Yang</name></author><author><name sortKey="Huang, Jia Bin" sort="Huang, Jia Bin" uniqKey="Huang J" first="Jia-Bin" last="Huang">Jia-Bin Huang</name></author></titleStmt><publicationStmt><idno type="wicri:source">PubMed</idno><date when="2020">2020</date><idno type="RBID">pubmed:32275584</idno><idno type="pmid">32275584</idno><idno type="doi">10.1109/TPAMI.2020.2985395</idno><idno type="wicri:Area/Main/Corpus">000204</idno><idno type="wicri:explorRef" wicri:stream="Main" wicri:step="Corpus" wicri:corpus="PubMed">000204</idno><idno type="wicri:Area/Main/Curation">000204</idno><idno type="wicri:explorRef" wicri:stream="Main" wicri:step="Curation">000204</idno><idno type="wicri:Area/Main/Exploration">000204</idno></publicationStmt><sourceDesc><biblStruct><analytic><title xml:lang="en">Show, Match and Segment: Joint Weakly Supervised Learning of Semantic Matching and Object Co-segmentation.</title><author><name sortKey="Chen, Yun Chun" sort="Chen, Yun Chun" uniqKey="Chen Y" first="Yun-Chun" last="Chen">Yun-Chun Chen</name></author><author><name sortKey="Lin, Yen Yu" sort="Lin, Yen Yu" uniqKey="Lin Y" first="Yen-Yu" last="Lin">Yen-Yu Lin</name></author><author><name sortKey="Yang, Ming Hsuan" sort="Yang, Ming Hsuan" uniqKey="Yang M" first="Ming-Hsuan" last="Yang">Ming-Hsuan Yang</name></author><author><name sortKey="Huang, Jia Bin" sort="Huang, Jia Bin" uniqKey="Huang J" first="Jia-Bin" last="Huang">Jia-Bin Huang</name></author></analytic><series><title level="j">IEEE transactions on pattern analysis and machine intelligence</title><idno type="eISSN">1939-3539</idno><imprint><date when="2020" type="published">2020</date></imprint></series></biblStruct></sourceDesc></fileDesc><profileDesc><textClass></textClass></profileDesc></teiHeader><front><div type="abstract" xml:lang="en">We present an approach for jointly matching and segmenting object instances of the same category within a collection of images. In contrast to existing algorithms that tackle the tasks of semantic matching and object co-segmentation in isolation, our method exploits the complementary nature of the two tasks. The key insights of our method are two-fold. First, the estimated dense correspondence fields from semantic matching provide supervision for object co-segmentation by enforcing consistency between the predicted masks from a pair of images. Second, the predicted object masks from object co-segmentation, in turn, allow us to reduce the adverse effects due to background clutters for improving semantic matching. Our model is end-to-end trainable and does not require supervision from manually annotated correspondences and object masks. We validate the efficacy of our approach on five benchmark datasets: TSS, Internet, PF-PASCAL, PF-WILLOW, and SPair-71k, and show that our algorithm performs favorably against the state-of-the-art methods on both semantic matching and object co-segmentation tasks.</div></front></TEI><pubmed><MedlineCitation Status="Publisher" Owner="NLM"><PMID Version="1">32275584</PMID><DateRevised><Year>2020</Year><Month>04</Month><Day>14</Day></DateRevised><Article PubModel="Print-Electronic"><Journal><ISSN IssnType="Electronic">1939-3539</ISSN><JournalIssue CitedMedium="Internet"><PubDate><Year>2020</Year><Month>Apr</Month><Day>06</Day></PubDate></JournalIssue><Title>IEEE transactions on pattern analysis and machine intelligence</Title><ISOAbbreviation>IEEE Trans Pattern Anal Mach Intell</ISOAbbreviation></Journal><ArticleTitle>Show, Match and Segment: Joint Weakly Supervised Learning of Semantic Matching and Object Co-segmentation.</ArticleTitle><ELocationID EIdType="doi" ValidYN="Y">10.1109/TPAMI.2020.2985395</ELocationID><Abstract><AbstractText>We present an approach for jointly matching and segmenting object instances of the same category within a collection of images. In contrast to existing algorithms that tackle the tasks of semantic matching and object co-segmentation in isolation, our method exploits the complementary nature of the two tasks. The key insights of our method are two-fold. First, the estimated dense correspondence fields from semantic matching provide supervision for object co-segmentation by enforcing consistency between the predicted masks from a pair of images. Second, the predicted object masks from object co-segmentation, in turn, allow us to reduce the adverse effects due to background clutters for improving semantic matching. Our model is end-to-end trainable and does not require supervision from manually annotated correspondences and object masks. We validate the efficacy of our approach on five benchmark datasets: TSS, Internet, PF-PASCAL, PF-WILLOW, and SPair-71k, and show that our algorithm performs favorably against the state-of-the-art methods on both semantic matching and object co-segmentation tasks.</AbstractText></Abstract><AuthorList CompleteYN="Y"><Author ValidYN="Y"><LastName>Chen</LastName><ForeName>Yun-Chun</ForeName><Initials>YC</Initials></Author><Author ValidYN="Y"><LastName>Lin</LastName><ForeName>Yen-Yu</ForeName><Initials>YY</Initials></Author><Author ValidYN="Y"><LastName>Yang</LastName><ForeName>Ming-Hsuan</ForeName><Initials>MH</Initials></Author><Author ValidYN="Y"><LastName>Huang</LastName><ForeName>Jia-Bin</ForeName><Initials>JB</Initials></Author></AuthorList><Language>eng</Language><PublicationTypeList><PublicationType UI="D016428">Journal Article</PublicationType></PublicationTypeList><ArticleDate DateType="Electronic"><Year>2020</Year><Month>04</Month><Day>06</Day></ArticleDate></Article><MedlineJournalInfo><Country>United States</Country><MedlineTA>IEEE Trans Pattern Anal Mach Intell</MedlineTA><NlmUniqueID>9885960</NlmUniqueID><ISSNLinking>0098-5589</ISSNLinking></MedlineJournalInfo><CitationSubset>IM</CitationSubset></MedlineCitation><PubmedData><History><PubMedPubDate PubStatus="pubmed"><Year>2020</Year><Month>4</Month><Day>11</Day><Hour>6</Hour><Minute>0</Minute></PubMedPubDate><PubMedPubDate PubStatus="medline"><Year>2020</Year><Month>4</Month><Day>11</Day><Hour>6</Hour><Minute>0</Minute></PubMedPubDate><PubMedPubDate PubStatus="entrez"><Year>2020</Year><Month>4</Month><Day>11</Day><Hour>6</Hour><Minute>0</Minute></PubMedPubDate></History><PublicationStatus>aheadofprint</PublicationStatus><ArticleIdList><ArticleId IdType="pubmed">32275584</ArticleId><ArticleId IdType="doi">10.1109/TPAMI.2020.2985395</ArticleId></ArticleIdList></PubmedData></pubmed><affiliations><list></list><tree><noCountry><name sortKey="Chen, Yun Chun" sort="Chen, Yun Chun" uniqKey="Chen Y" first="Yun-Chun" last="Chen">Yun-Chun Chen</name><name sortKey="Huang, Jia Bin" sort="Huang, Jia Bin" uniqKey="Huang J" first="Jia-Bin" last="Huang">Jia-Bin Huang</name><name sortKey="Lin, Yen Yu" sort="Lin, Yen Yu" uniqKey="Lin Y" first="Yen-Yu" last="Lin">Yen-Yu Lin</name><name sortKey="Yang, Ming Hsuan" sort="Yang, Ming Hsuan" uniqKey="Yang M" first="Ming-Hsuan" last="Yang">Ming-Hsuan Yang</name></noCountry></tree></affiliations></record>Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Bois/explor/WillowV1/Data/Main/Exploration
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 000077 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Main/Exploration/biblio.hfd -nk 000077 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien
|wiki= Bois
|area= WillowV1
|flux= Main
|étape= Exploration
|type= RBID
|clé= pubmed:32275584
|texte= Show, Match and Segment: Joint Weakly Supervised Learning of Semantic Matching and Object Co-segmentation.
}}
Pour générer des pages wiki
HfdIndexSelect -h $EXPLOR_AREA/Data/Main/Exploration/RBID.i -Sk "pubmed:32275584" \
| HfdSelect -Kh $EXPLOR_AREA/Data/Main/Exploration/biblio.hfd \
| NlmPubMed2Wicri -a WillowV1
| This area was generated with Dilib version V0.6.37. |  |