<?xml version="1.0" encoding="utf-8"?>
<TEI xmlns="http://www.tei-c.org/ns/1.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:hal="http://hal.archives-ouvertes.fr/" xmlns:gml="http://www.opengis.net/gml/3.3/" xmlns:gmlce="http://www.opengis.net/gml/3.3/ce" version="1.1" xsi:schemaLocation="http://www.tei-c.org/ns/1.0 http://api.archives-ouvertes.fr/documents/aofr-sword.xsd">
  <teiHeader>
    <fileDesc>
      <titleStmt>
        <title>HAL TEI export of hal-01428963v2</title>
      </titleStmt>
      <publicationStmt>
        <distributor>CCSD</distributor>
        <availability status="restricted">
          <licence target="https://creativecommons.org/publicdomain/zero/1.0/">CC0 1.0 - Universal</licence>
        </availability>
        <date when="2026-05-22T03:50:30+02:00"/>
      </publicationStmt>
      <sourceDesc>
        <p part="N">HAL API Platform</p>
      </sourceDesc>
    </fileDesc>
  </teiHeader>
  <text>
    <body>
      <listBibl>
        <biblFull>
          <titleStmt>
            <title xml:lang="en">Areas of Attention for Image Captioning</title>
            <author role="aut">
              <persName>
                <forename type="first">Marco</forename>
                <surname>Pedersoli</surname>
              </persName>
              <idno type="halauthorid">1101413-0</idno>
              <affiliation ref="#struct-136804"/>
            </author>
            <author role="aut">
              <persName>
                <forename type="first">Thomas</forename>
                <surname>Lucas</surname>
              </persName>
              <idno type="halauthorid">1101414-0</idno>
              <affiliation ref="#struct-445108"/>
            </author>
            <author role="aut">
              <persName>
                <forename type="first">Cordelia</forename>
                <surname>Schmid</surname>
              </persName>
              <idno type="halauthorid">119565-0</idno>
              <affiliation ref="#struct-445108"/>
            </author>
            <author role="aut">
              <persName>
                <forename type="first">Jakob</forename>
                <surname>Verbeek</surname>
              </persName>
              <email type="md5">fba6494e81069c5dae86c46d805756c3</email>
              <email type="domain">gmail.com</email>
              <idno type="idhal" notation="string">verbeek</idno>
              <idno type="idhal" notation="numeric">10676</idno>
              <idno type="halauthorid" notation="string">23230-10676</idno>
              <idno type="ORCID">https://orcid.org/0000-0003-1419-1816</idno>
              <idno type="GOOGLE SCHOLAR">https://scholar.google.com/citations?user=oZGA-rAAAAAJ&amp;hl=en</idno>
              <idno type="IDREF">https://www.idref.fr/180998463</idno>
              <affiliation ref="#struct-445108"/>
            </author>
            <editor role="depositor">
              <persName>
                <forename>THOTH</forename>
                <surname>Team</surname>
              </persName>
              <email type="md5">413829dab28a0224ec5dc6d17391bece</email>
              <email type="domain">inria.fr</email>
            </editor>
            <funder ref="#projanr-38053"/>
            <funder ref="#projanr-42147"/>
            <funder ref="#projeurop-98942"/>
          </titleStmt>
          <editionStmt>
            <edition n="v1">
              <date type="whenSubmitted">2017-01-06 16:42:08</date>
            </edition>
            <edition n="v2" type="current">
              <date type="whenSubmitted">2017-08-25 16:06:00</date>
              <date type="whenModified">2026-04-23 09:02:01</date>
              <date type="whenReleased">2017-08-25 17:31:01</date>
              <date type="whenProduced">2017-10-22</date>
              <date type="whenEndEmbargoed">2017-08-25</date>
              <ref type="file" target="https://inria.hal.science/hal-01428963v2/document">
                <date notBefore="2017-08-25"/>
              </ref>
              <ref type="file" subtype="author" n="1" target="https://inria.hal.science/hal-01428963v2/file/pedersoli17iccv.pdf" id="file-1577424-1630714">
                <date notBefore="2017-08-25"/>
              </ref>
              <ref type="externalLink" target="http://arxiv.org/pdf/1612.01033"/>
            </edition>
            <respStmt>
              <resp>contributor</resp>
              <name key="156804">
                <persName>
                  <forename>THOTH</forename>
                  <surname>Team</surname>
                </persName>
                <email type="md5">413829dab28a0224ec5dc6d17391bece</email>
                <email type="domain">inria.fr</email>
              </name>
            </respStmt>
          </editionStmt>
          <publicationStmt>
            <distributor>CCSD</distributor>
            <idno type="halId">hal-01428963</idno>
            <idno type="halUri">https://inria.hal.science/hal-01428963</idno>
            <idno type="halBibtex">pedersoli:hal-01428963</idno>
            <idno type="halRefHtml">&lt;i&gt;ICCV - International Conference on Computer Vision&lt;/i&gt;, Oct 2017, Venice, Italy. pp.1251-1259, &lt;a target="_blank" href="https://dx.doi.org/10.1109/ICCV.2017.140"&gt;&amp;#x27E8;10.1109/ICCV.2017.140&amp;#x27E9;&lt;/a&gt;</idno>
            <idno type="halRef">ICCV - International Conference on Computer Vision, Oct 2017, Venice, Italy. pp.1251-1259, &amp;#x27E8;10.1109/ICCV.2017.140&amp;#x27E9;</idno>
            <availability status="restricted">
              <licence target="https://about.hal.science/hal-authorisation-v1/">HAL Authorization<ref corresp="#file-1577424-1630714"/></licence>
            </availability>
          </publicationStmt>
          <seriesStmt>
            <idno type="stamp" n="UGA">HAL Grenoble Alpes</idno>
            <idno type="stamp" n="CNRS">CNRS - Centre national de la recherche scientifique</idno>
            <idno type="stamp" n="INRIA">INRIA - Institut National de Recherche en Informatique et en Automatique</idno>
            <idno type="stamp" n="INPG">Institut polytechnique de Grenoble</idno>
            <idno type="stamp" n="INRIA-RHA">INRIA Grenoble - Rhône-Alpes</idno>
            <idno type="stamp" n="INSMI">CNRS-INSMI - INstitut des Sciences Mathématiques et de leurs Interactions</idno>
            <idno type="stamp" n="OPENAIRE">OpenAIRE</idno>
            <idno type="stamp" n="INRIA_TEST">INRIA - Institut National de Recherche en Informatique et en Automatique</idno>
            <idno type="stamp" n="LJK">Laboratoire Jean Kuntzmann</idno>
            <idno type="stamp" n="LJK_GI" corresp="LJK">Département Géométrie et Images</idno>
            <idno type="stamp" n="TESTALAIN1">TESTALAIN1</idno>
            <idno type="stamp" n="PERSYVAL-LAB">[Labex] PERSYVAL-lab</idno>
            <idno type="stamp" n="INRIA2">INRIA 2</idno>
            <idno type="stamp" n="LJK-GI-THOTH" corresp="LJK_GI">Apprentissage de modèles visuels à partir de données massives</idno>
            <idno type="stamp" n="INRIA2017">INRIA2017</idno>
            <idno type="stamp" n="INRIA-RENGRE">INRIA-RENGRE</idno>
            <idno type="stamp" n="UGA-COMUE">Université Grenoble Alpes [2016-2019]</idno>
            <idno type="stamp" n="ANR">ANR</idno>
            <idno type="stamp" n="INRIA-CANADA">INRIA-CANADA</idno>
            <idno type="stamp" n="TEST-UGA">TEST-UGA</idno>
          </seriesStmt>
          <notesStmt>
            <note type="audience" n="2">International</note>
            <note type="invited" n="0">No</note>
            <note type="popular" n="0">No</note>
            <note type="peer" n="1">Yes</note>
            <note type="proceedings" n="1">Yes</note>
          </notesStmt>
          <sourceDesc>
            <biblStruct>
              <analytic>
                <title xml:lang="en">Areas of Attention for Image Captioning</title>
                <author role="aut">
                  <persName>
                    <forename type="first">Marco</forename>
                    <surname>Pedersoli</surname>
                  </persName>
                  <idno type="halauthorid">1101413-0</idno>
                  <affiliation ref="#struct-136804"/>
                </author>
                <author role="aut">
                  <persName>
                    <forename type="first">Thomas</forename>
                    <surname>Lucas</surname>
                  </persName>
                  <idno type="halauthorid">1101414-0</idno>
                  <affiliation ref="#struct-445108"/>
                </author>
                <author role="aut">
                  <persName>
                    <forename type="first">Cordelia</forename>
                    <surname>Schmid</surname>
                  </persName>
                  <idno type="halauthorid">119565-0</idno>
                  <affiliation ref="#struct-445108"/>
                </author>
                <author role="aut">
                  <persName>
                    <forename type="first">Jakob</forename>
                    <surname>Verbeek</surname>
                  </persName>
                  <email type="md5">fba6494e81069c5dae86c46d805756c3</email>
                  <email type="domain">gmail.com</email>
                  <idno type="idhal" notation="string">verbeek</idno>
                  <idno type="idhal" notation="numeric">10676</idno>
                  <idno type="halauthorid" notation="string">23230-10676</idno>
                  <idno type="ORCID">https://orcid.org/0000-0003-1419-1816</idno>
                  <idno type="GOOGLE SCHOLAR">https://scholar.google.com/citations?user=oZGA-rAAAAAJ&amp;hl=en</idno>
                  <idno type="IDREF">https://www.idref.fr/180998463</idno>
                  <affiliation ref="#struct-445108"/>
                </author>
              </analytic>
              <monogr>
                <meeting>
                  <title>ICCV - International Conference on Computer Vision</title>
                  <date type="start">2017-10-22</date>
                  <date type="end">2017-10-29</date>
                  <settlement>Venice</settlement>
                  <country key="IT">Italy</country>
                </meeting>
                <imprint>
                  <publisher>IEEE</publisher>
                  <biblScope unit="pp">1251-1259</biblScope>
                  <date type="datePub">2017-12</date>
                </imprint>
              </monogr>
              <idno type="doi">10.1109/ICCV.2017.140</idno>
            </biblStruct>
          </sourceDesc>
          <profileDesc>
            <langUsage>
              <language ident="en">English</language>
            </langUsage>
            <textClass>
              <classCode scheme="halDomain" n="info.info-cv">Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV]</classCode>
              <classCode scheme="halTypology" n="COMM">Conference papers</classCode>
              <classCode scheme="halOldTypology" n="COMM">Conference papers</classCode>
              <classCode scheme="halTreeTypology" n="COMM">Conference papers</classCode>
            </textClass>
            <abstract xml:lang="en">
              <p>We propose ``Areas of Attention'', a novel attention-based model for automatic image captioning. Our approach models  the dependencies between image regions, caption words, and the state of an RNN language model, using three pairwise interactions. In contrast to previous attention-based approaches that associate image regions only to the RNN state, our method allows a direct association between caption words and image regions. During training these associations are inferred from image-level captions, akin to weakly-supervised object detector training. These associations help to improve captioning by localizing the corresponding regions during testing. We also propose and compare different ways of generating attention areas: CNN activation grids, object proposals, and spatial transformers nets applied in a convolutional fashion. Spatial transformers give the best results. They allow for image specific attention areas, and can be trained jointly with the rest of the network. Our attention mechanism and spatial transformer attention areas together yield state-of-the-art results on the MSCOCO dataset.</p>
            </abstract>
            <particDesc>
              <org type="consortium">ERC_Allegro</org>
            </particDesc>
          </profileDesc>
        </biblFull>
      </listBibl>
    </body>
    <back>
      <listOrg type="structures">
        <org type="institution" xml:id="struct-136804" status="VALID">
          <idno type="IdRef">080094201</idno>
          <idno type="ISNI">0000000122224302</idno>
          <idno type="ROR">https://ror.org/0020snb74</idno>
          <idno type="Wikidata">Q3577916</idno>
          <orgName>École de technologie supérieure [Montréal]</orgName>
          <orgName type="acronym">ÉTS</orgName>
          <date type="start">1974-01-01</date>
          <desc>
            <address>
              <addrLine>1100, rue Notre-Dame Ouest Montréal (Qc) H3C 1K3 Canada</addrLine>
              <country key="CA"/>
            </address>
            <ref type="url">http://www.etsmtl.ca/</ref>
          </desc>
        </org>
        <org type="researchteam" xml:id="struct-445108" status="OLD">
          <idno type="IdRef">242196691</idno>
          <idno type="RNSR">201622034K</idno>
          <orgName>Apprentissage de modèles à partir de données massives</orgName>
          <orgName type="acronym">Thoth</orgName>
          <date type="start">2016-01-01</date>
          <date type="end">2019-12-31</date>
          <desc>
            <address>
              <country key="FR"/>
            </address>
            <ref type="url">http://www.inria.fr/equipes/thoth</ref>
          </desc>
          <listRelation>
            <relation active="#struct-2497" type="direct"/>
            <relation active="#struct-300009" type="indirect"/>
            <relation active="#struct-1042161" type="direct"/>
            <relation active="#struct-89889" type="indirect"/>
            <relation name="UMR5224" active="#struct-441569" type="indirect"/>
            <relation active="#struct-445543" type="indirect"/>
          </listRelation>
        </org>
        <org type="laboratory" xml:id="struct-2497" status="VALID">
          <idno type="RNSR">199218244V</idno>
          <idno type="ROR">https://ror.org/00n8d6z93</idno>
          <orgName>Centre Inria de l'Université Grenoble Alpes</orgName>
          <desc>
            <address>
              <addrLine>Inovallée655 avenue de l'Europe38330 Montbonnot</addrLine>
              <country key="FR"/>
            </address>
            <ref type="url">http://www.inria.fr/centre/grenoble</ref>
          </desc>
          <listRelation>
            <relation active="#struct-300009" type="direct"/>
          </listRelation>
        </org>
        <org type="institution" xml:id="struct-300009" status="VALID">
          <idno type="ROR">https://ror.org/02kvxyf05</idno>
          <orgName>Institut National de Recherche en Informatique et en Automatique</orgName>
          <orgName type="acronym">Inria</orgName>
          <desc>
            <address>
              <addrLine>Domaine de VoluceauRocquencourt - BP 10578153 Le Chesnay Cedex</addrLine>
              <country key="FR"/>
            </address>
            <ref type="url">http://www.inria.fr/en/</ref>
          </desc>
        </org>
        <org type="laboratory" xml:id="struct-1042161" status="OLD">
          <idno type="IdRef">184945011</idno>
          <idno type="ISNI">000000040383676X</idno>
          <idno type="RNSR">200711891Z</idno>
          <idno type="ROR">https://ror.org/04ett5b41</idno>
          <orgName>Laboratoire Jean Kuntzmann</orgName>
          <orgName type="acronym">LJK</orgName>
          <date type="start">2016-01-01</date>
          <date type="end">2019-12-31</date>
          <desc>
            <address>
              <addrLine>Bâtiment IMAG, CS 40700, F-38058 Grenoble Cedex 9</addrLine>
              <country key="FR"/>
            </address>
            <ref type="url">https://ljk.imag.fr/</ref>
          </desc>
          <listRelation>
            <relation active="#struct-89889" type="direct"/>
            <relation active="#struct-300009" type="direct"/>
            <relation name="UMR5224" active="#struct-441569" type="direct"/>
            <relation active="#struct-445543" type="direct"/>
          </listRelation>
        </org>
        <org type="institution" xml:id="struct-89889" status="OLD">
          <idno type="IdRef">026388804</idno>
          <idno type="ROR">https://ror.org/05sbt2524</idno>
          <orgName>Institut polytechnique de Grenoble - Grenoble Institute of Technology</orgName>
          <orgName type="acronym">Grenoble INP</orgName>
          <date type="start">2007-01-01</date>
          <date type="end">2019-12-31</date>
          <desc>
            <address>
              <addrLine>46 avenue Félix Viallet 38031 Grenoble Cedex 1</addrLine>
              <country key="FR"/>
            </address>
            <ref type="url">http://www.grenoble-inp.fr/</ref>
          </desc>
        </org>
        <org type="regroupinstitution" xml:id="struct-441569" status="VALID">
          <idno type="IdRef">02636817X</idno>
          <idno type="ISNI">0000000122597504</idno>
          <idno type="ROR">https://ror.org/02feahw73</idno>
          <orgName>Centre National de la Recherche Scientifique</orgName>
          <orgName type="acronym">CNRS</orgName>
          <date type="start">1939-10-19</date>
          <desc>
            <address>
              <country key="FR"/>
            </address>
            <ref type="url">https://www.cnrs.fr/</ref>
          </desc>
        </org>
        <org type="institution" xml:id="struct-445543" status="OLD">
          <idno type="IdRef">188399275</idno>
          <idno type="ROR">https://ror.org/02rx3b187</idno>
          <orgName>Université Grenoble Alpes [2016-2019]</orgName>
          <orgName type="acronym">UGA [2016-2019]</orgName>
          <date type="start">2016-01-01</date>
          <date type="end">2019-12-31</date>
          <desc>
            <address>
              <addrLine>38058 Grenoble cedex</addrLine>
              <country key="FR"/>
            </address>
          </desc>
        </org>
      </listOrg>
      <listOrg type="projects">
        <org type="anrProject" xml:id="projanr-38053" status="VALID">
          <idno type="anr">ANR-11-LABX-0025</idno>
          <idno type="program">Laboratoires d'excellence</idno>
          <orgName>PERSYVAL-lab</orgName>
          <desc>Systemes et Algorithmes Pervasifs au confluent des mondes physique et numérique</desc>
          <date type="start">2011</date>
        </org>
        <org type="anrProject" xml:id="projanr-42147" status="VALID">
          <idno type="anr">ANR-16-CE23-0006</idno>
          <orgName>Deep_in_France</orgName>
          <desc>Réseaux de neurones profonds pour l'apprentissage</desc>
          <date type="start">2016</date>
        </org>
        <org type="europeanProject" xml:id="projeurop-98942" status="VALID">
          <idno type="number">320559</idno>
          <idno type="program">ERC-2012-ADG_20120216</idno>
          <idno type="call">ERC-2012-ADG_20120216</idno>
          <orgName>ALLEGRO</orgName>
          <desc>Active large-scale learning for visual recognition</desc>
          <date type="start">2013-01-04</date>
          <date type="end">2019-03-31</date>
        </org>
      </listOrg>
    </back>
  </text>
</TEI>