Visual Affinity Propagation Improves Sub-topics Diversity without Loss of Precision in Web Photo Retrieval View Full Text


Ontology type: schema:Chapter     


Chapter Info

DATE

2009

AUTHORS

Hervé Glotin , Zhong-Qiu Zhao

ABSTRACT

This paper demonstrates that Affinity Propagation (AP) outperforms Kmeans for sub-topic clustering of web image retrieval. A SVM visual images retrieval system is built, and then clustering is performed on the results of each topic. Then we heighten the diversity of the 20 top results, by moving into the top the image with the lowest rank in each cluster. Using 45 dimensions Profile Entropy visual Features, we show for the 39 topics of the imageCLEF08 web image retrieval clustering campaign on 20K IAPR images, that the Cluster-Recall (CR) after AP is 13% better than the baseline without clustering, while the Precision stays almost the same. Moreover, CR and Precision without clustering are altered by Kmeans. We finally discuss that some high-level topics require text information for good CR, and that more discriminant visual features would also allow Precision enhancement after AP. More... »

PAGES

628-631

References to SciGraph publications

  • 2009. A Comparative Study of Diversity Methods for Hybrid Text and Image Retrieval Approaches in EVALUATING SYSTEMS FOR MULTILINGUAL AND MULTIMODAL INFORMATION ACCESS
  • Book

    TITLE

    Evaluating Systems for Multilingual and Multimodal Information Access

    ISBN

    978-3-642-04446-5
    978-3-642-04447-2

    Author Affiliations

    Identifiers

    URI

    http://scigraph.springernature.com/pub.10.1007/978-3-642-04447-2_78

    DOI

    http://dx.doi.org/10.1007/978-3-642-04447-2_78

    DIMENSIONS

    https://app.dimensions.ai/details/publication/pub.1036898399


    Indexing Status Check whether this publication has been indexed by Scopus and Web Of Science using the SN Indexing Status Tool
    Incoming Citations Browse incoming citations for this publication using opencitations.net

    JSON-LD is the canonical representation for SciGraph data.

    TIP: You can open this SciGraph record using an external JSON-LD service: JSON-LD Playground Google SDTT

    [
      {
        "@context": "https://springernature.github.io/scigraph/jsonld/sgcontext.json", 
        "about": [
          {
            "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/0801", 
            "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
            "name": "Artificial Intelligence and Image Processing", 
            "type": "DefinedTerm"
          }, 
          {
            "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/08", 
            "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
            "name": "Information and Computing Sciences", 
            "type": "DefinedTerm"
          }
        ], 
        "author": [
          {
            "affiliation": {
              "alternateName": "Universite De Toulon Et Du Var", 
              "id": "https://www.grid.ac/institutes/grid.12611.35", 
              "name": [
                "Systems & Information Sciences Lab., UMR CNRS 6168, & Univ. Sud Toulon-Var, France"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Glotin", 
            "givenName": "Herv\u00e9", 
            "id": "sg:person.016622300103.82", 
            "sameAs": [
              "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.016622300103.82"
            ], 
            "type": "Person"
          }, 
          {
            "affiliation": {
              "alternateName": "Universite De Toulon Et Du Var", 
              "id": "https://www.grid.ac/institutes/grid.12611.35", 
              "name": [
                "Systems & Information Sciences Lab., UMR CNRS 6168, & Univ. Sud Toulon-Var, France"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Zhao", 
            "givenName": "Zhong-Qiu", 
            "id": "sg:person.016102547154.53", 
            "sameAs": [
              "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.016102547154.53"
            ], 
            "type": "Person"
          }
        ], 
        "citation": [
          {
            "id": "https://doi.org/10.1126/science.1136800", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1017347292"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1007/978-3-642-04447-2_72", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1052833888", 
              "https://doi.org/10.1007/978-3-642-04447-2_72"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "https://doi.org/10.1109/icassp.2008.4517838", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1095351518"
            ], 
            "type": "CreativeWork"
          }
        ], 
        "datePublished": "2009", 
        "datePublishedReg": "2009-01-01", 
        "description": "This paper demonstrates that Affinity Propagation (AP) outperforms Kmeans for sub-topic clustering of web image retrieval. A SVM visual images retrieval system is built, and then clustering is performed on the results of each topic. Then we heighten the diversity of the 20 top results, by moving into the top the image with the lowest rank in each cluster. Using 45 dimensions Profile Entropy visual Features, we show for the 39 topics of the imageCLEF08 web image retrieval clustering campaign on 20K IAPR images, that the Cluster-Recall (CR) after AP is 13% better than the baseline without clustering, while the Precision stays almost the same. Moreover, CR and Precision without clustering are altered by Kmeans. We finally discuss that some high-level topics require text information for good CR, and that more discriminant visual features would also allow Precision enhancement after AP.", 
        "editor": [
          {
            "familyName": "Peters", 
            "givenName": "Carol", 
            "type": "Person"
          }, 
          {
            "familyName": "Deselaers", 
            "givenName": "Thomas", 
            "type": "Person"
          }, 
          {
            "familyName": "Ferro", 
            "givenName": "Nicola", 
            "type": "Person"
          }, 
          {
            "familyName": "Gonzalo", 
            "givenName": "Julio", 
            "type": "Person"
          }, 
          {
            "familyName": "Jones", 
            "givenName": "Gareth J. F.", 
            "type": "Person"
          }, 
          {
            "familyName": "Kurimo", 
            "givenName": "Mikko", 
            "type": "Person"
          }, 
          {
            "familyName": "Mandl", 
            "givenName": "Thomas", 
            "type": "Person"
          }, 
          {
            "familyName": "Pe\u00f1as", 
            "givenName": "Anselmo", 
            "type": "Person"
          }, 
          {
            "familyName": "Petras", 
            "givenName": "Vivien", 
            "type": "Person"
          }
        ], 
        "genre": "chapter", 
        "id": "sg:pub.10.1007/978-3-642-04447-2_78", 
        "inLanguage": [
          "en"
        ], 
        "isAccessibleForFree": false, 
        "isPartOf": {
          "isbn": [
            "978-3-642-04446-5", 
            "978-3-642-04447-2"
          ], 
          "name": "Evaluating Systems for Multilingual and Multimodal Information Access", 
          "type": "Book"
        }, 
        "name": "Visual Affinity Propagation Improves Sub-topics Diversity without Loss of Precision in Web Photo Retrieval", 
        "pagination": "628-631", 
        "productId": [
          {
            "name": "doi", 
            "type": "PropertyValue", 
            "value": [
              "10.1007/978-3-642-04447-2_78"
            ]
          }, 
          {
            "name": "readcube_id", 
            "type": "PropertyValue", 
            "value": [
              "afc5163698046e81b3901c9db3f2c67d2c54a2a716b08bca6e9888f5b58be5c2"
            ]
          }, 
          {
            "name": "dimensions_id", 
            "type": "PropertyValue", 
            "value": [
              "pub.1036898399"
            ]
          }
        ], 
        "publisher": {
          "location": "Berlin, Heidelberg", 
          "name": "Springer Berlin Heidelberg", 
          "type": "Organisation"
        }, 
        "sameAs": [
          "https://doi.org/10.1007/978-3-642-04447-2_78", 
          "https://app.dimensions.ai/details/publication/pub.1036898399"
        ], 
        "sdDataset": "chapters", 
        "sdDatePublished": "2019-04-15T22:00", 
        "sdLicense": "https://scigraph.springernature.com/explorer/license/", 
        "sdPublisher": {
          "name": "Springer Nature - SN SciGraph project", 
          "type": "Organization"
        }, 
        "sdSource": "s3://com-uberresearch-data-dimensions-target-20181106-alternative/cleanup/v134/2549eaecd7973599484d7c17b260dba0a4ecb94b/merge/v9/a6c9fde33151104705d4d7ff012ea9563521a3ce/jats-lookup/v90/0000000001_0000000264/records_8693_00000266.jsonl", 
        "type": "Chapter", 
        "url": "http://link.springer.com/10.1007/978-3-642-04447-2_78"
      }
    ]
     

    Download the RDF metadata as:  json-ld nt turtle xml License info

    HOW TO GET THIS DATA PROGRAMMATICALLY:

    JSON-LD is a popular format for linked data which is fully compatible with JSON.

    curl -H 'Accept: application/ld+json' 'https://scigraph.springernature.com/pub.10.1007/978-3-642-04447-2_78'

    N-Triples is a line-based linked data format ideal for batch operations.

    curl -H 'Accept: application/n-triples' 'https://scigraph.springernature.com/pub.10.1007/978-3-642-04447-2_78'

    Turtle is a human-readable linked data format.

    curl -H 'Accept: text/turtle' 'https://scigraph.springernature.com/pub.10.1007/978-3-642-04447-2_78'

    RDF/XML is a standard XML format for linked data.

    curl -H 'Accept: application/rdf+xml' 'https://scigraph.springernature.com/pub.10.1007/978-3-642-04447-2_78'


     

    This table displays all metadata directly associated to this object as RDF triples.

    122 TRIPLES      23 PREDICATES      30 URIs      20 LITERALS      8 BLANK NODES

    Subject Predicate Object
    1 sg:pub.10.1007/978-3-642-04447-2_78 schema:about anzsrc-for:08
    2 anzsrc-for:0801
    3 schema:author N1b0d279c14c44fa3a9b863727c9844f7
    4 schema:citation sg:pub.10.1007/978-3-642-04447-2_72
    5 https://doi.org/10.1109/icassp.2008.4517838
    6 https://doi.org/10.1126/science.1136800
    7 schema:datePublished 2009
    8 schema:datePublishedReg 2009-01-01
    9 schema:description This paper demonstrates that Affinity Propagation (AP) outperforms Kmeans for sub-topic clustering of web image retrieval. A SVM visual images retrieval system is built, and then clustering is performed on the results of each topic. Then we heighten the diversity of the 20 top results, by moving into the top the image with the lowest rank in each cluster. Using 45 dimensions Profile Entropy visual Features, we show for the 39 topics of the imageCLEF08 web image retrieval clustering campaign on 20K IAPR images, that the Cluster-Recall (CR) after AP is 13% better than the baseline without clustering, while the Precision stays almost the same. Moreover, CR and Precision without clustering are altered by Kmeans. We finally discuss that some high-level topics require text information for good CR, and that more discriminant visual features would also allow Precision enhancement after AP.
    10 schema:editor N31d4a9ac5f7842878f5d4a8e3234d63a
    11 schema:genre chapter
    12 schema:inLanguage en
    13 schema:isAccessibleForFree false
    14 schema:isPartOf Ne5d3ebf97e2b466c83cf98c9022bcbc2
    15 schema:name Visual Affinity Propagation Improves Sub-topics Diversity without Loss of Precision in Web Photo Retrieval
    16 schema:pagination 628-631
    17 schema:productId N105f217dd82b499a9275c96369356048
    18 N17fb2bc743934f2eac224104cf407c81
    19 N78ae1dfcb8844d87bd3058ad8dc46fd6
    20 schema:publisher Nc74316b1ad1e4bb5b37f1c674989d26f
    21 schema:sameAs https://app.dimensions.ai/details/publication/pub.1036898399
    22 https://doi.org/10.1007/978-3-642-04447-2_78
    23 schema:sdDatePublished 2019-04-15T22:00
    24 schema:sdLicense https://scigraph.springernature.com/explorer/license/
    25 schema:sdPublisher Nc233d963d2814712854335291b377a0b
    26 schema:url http://link.springer.com/10.1007/978-3-642-04447-2_78
    27 sgo:license sg:explorer/license/
    28 sgo:sdDataset chapters
    29 rdf:type schema:Chapter
    30 N0420133eba474cde9fe3f14f49956dcd schema:familyName Peters
    31 schema:givenName Carol
    32 rdf:type schema:Person
    33 N0f9cdc85477f4e75882bf32eebc782eb rdf:first N31d24c475aa143f9b1347e76005c68aa
    34 rdf:rest Na9daa83a8a1e40218a591796d7bf3fc7
    35 N105f217dd82b499a9275c96369356048 schema:name dimensions_id
    36 schema:value pub.1036898399
    37 rdf:type schema:PropertyValue
    38 N17fb2bc743934f2eac224104cf407c81 schema:name doi
    39 schema:value 10.1007/978-3-642-04447-2_78
    40 rdf:type schema:PropertyValue
    41 N1b0d279c14c44fa3a9b863727c9844f7 rdf:first sg:person.016622300103.82
    42 rdf:rest N6936b16fd96a45be8312707663a03eea
    43 N1fbce4e07a3f4f4687e52947c1b97f1d rdf:first N73f6f95978ee412a8d773e7365d9cd62
    44 rdf:rest N73c2988161e24f7fb76b0fac013bfe1c
    45 N31d24c475aa143f9b1347e76005c68aa schema:familyName Jones
    46 schema:givenName Gareth J. F.
    47 rdf:type schema:Person
    48 N31d4a9ac5f7842878f5d4a8e3234d63a rdf:first N0420133eba474cde9fe3f14f49956dcd
    49 rdf:rest Nf04c471d99f648f6a66de1ba79b06c74
    50 N4fbf6c39204c4f408a7143b972c42213 schema:familyName Petras
    51 schema:givenName Vivien
    52 rdf:type schema:Person
    53 N51031d71aedd41889f4ad06c55896c22 schema:familyName Mandl
    54 schema:givenName Thomas
    55 rdf:type schema:Person
    56 N5c9b3cea9b434288ab3f7e0bd3b9dc85 rdf:first N73611740ff2e49e8bd9eeb253f279da6
    57 rdf:rest N5f3410fc44e74f228c6ab7f84bda9a06
    58 N5f3410fc44e74f228c6ab7f84bda9a06 rdf:first N4fbf6c39204c4f408a7143b972c42213
    59 rdf:rest rdf:nil
    60 N62ac5fadacc64d35943ce08df406085b schema:familyName Kurimo
    61 schema:givenName Mikko
    62 rdf:type schema:Person
    63 N6936b16fd96a45be8312707663a03eea rdf:first sg:person.016102547154.53
    64 rdf:rest rdf:nil
    65 N73611740ff2e49e8bd9eeb253f279da6 schema:familyName Peñas
    66 schema:givenName Anselmo
    67 rdf:type schema:Person
    68 N73c2988161e24f7fb76b0fac013bfe1c rdf:first Nc85378296b2f4b72a605a8298e1c0f39
    69 rdf:rest N0f9cdc85477f4e75882bf32eebc782eb
    70 N73f6f95978ee412a8d773e7365d9cd62 schema:familyName Ferro
    71 schema:givenName Nicola
    72 rdf:type schema:Person
    73 N78ae1dfcb8844d87bd3058ad8dc46fd6 schema:name readcube_id
    74 schema:value afc5163698046e81b3901c9db3f2c67d2c54a2a716b08bca6e9888f5b58be5c2
    75 rdf:type schema:PropertyValue
    76 N8a82217d70b240b5b422f35700fe6711 rdf:first N51031d71aedd41889f4ad06c55896c22
    77 rdf:rest N5c9b3cea9b434288ab3f7e0bd3b9dc85
    78 Na9daa83a8a1e40218a591796d7bf3fc7 rdf:first N62ac5fadacc64d35943ce08df406085b
    79 rdf:rest N8a82217d70b240b5b422f35700fe6711
    80 Nc233d963d2814712854335291b377a0b schema:name Springer Nature - SN SciGraph project
    81 rdf:type schema:Organization
    82 Nc576829754ff490fa047ca42b9f3518b schema:familyName Deselaers
    83 schema:givenName Thomas
    84 rdf:type schema:Person
    85 Nc74316b1ad1e4bb5b37f1c674989d26f schema:location Berlin, Heidelberg
    86 schema:name Springer Berlin Heidelberg
    87 rdf:type schema:Organisation
    88 Nc85378296b2f4b72a605a8298e1c0f39 schema:familyName Gonzalo
    89 schema:givenName Julio
    90 rdf:type schema:Person
    91 Ne5d3ebf97e2b466c83cf98c9022bcbc2 schema:isbn 978-3-642-04446-5
    92 978-3-642-04447-2
    93 schema:name Evaluating Systems for Multilingual and Multimodal Information Access
    94 rdf:type schema:Book
    95 Nf04c471d99f648f6a66de1ba79b06c74 rdf:first Nc576829754ff490fa047ca42b9f3518b
    96 rdf:rest N1fbce4e07a3f4f4687e52947c1b97f1d
    97 anzsrc-for:08 schema:inDefinedTermSet anzsrc-for:
    98 schema:name Information and Computing Sciences
    99 rdf:type schema:DefinedTerm
    100 anzsrc-for:0801 schema:inDefinedTermSet anzsrc-for:
    101 schema:name Artificial Intelligence and Image Processing
    102 rdf:type schema:DefinedTerm
    103 sg:person.016102547154.53 schema:affiliation https://www.grid.ac/institutes/grid.12611.35
    104 schema:familyName Zhao
    105 schema:givenName Zhong-Qiu
    106 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.016102547154.53
    107 rdf:type schema:Person
    108 sg:person.016622300103.82 schema:affiliation https://www.grid.ac/institutes/grid.12611.35
    109 schema:familyName Glotin
    110 schema:givenName Hervé
    111 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.016622300103.82
    112 rdf:type schema:Person
    113 sg:pub.10.1007/978-3-642-04447-2_72 schema:sameAs https://app.dimensions.ai/details/publication/pub.1052833888
    114 https://doi.org/10.1007/978-3-642-04447-2_72
    115 rdf:type schema:CreativeWork
    116 https://doi.org/10.1109/icassp.2008.4517838 schema:sameAs https://app.dimensions.ai/details/publication/pub.1095351518
    117 rdf:type schema:CreativeWork
    118 https://doi.org/10.1126/science.1136800 schema:sameAs https://app.dimensions.ai/details/publication/pub.1017347292
    119 rdf:type schema:CreativeWork
    120 https://www.grid.ac/institutes/grid.12611.35 schema:alternateName Universite De Toulon Et Du Var
    121 schema:name Systems & Information Sciences Lab., UMR CNRS 6168, & Univ. Sud Toulon-Var, France
    122 rdf:type schema:Organization
     




    Preview window. Press ESC to close (or click here)


    ...