June 30, 2023 Dataset Open Access

Hussein Mohammed

MARC21 XML Export

<?xml version='1.0' encoding='UTF-8'?>
<record xmlns="http://www.loc.gov/MARC21/slim">
  <leader>00000nmm##2200000uu#4500</leader>
  <datafield tag="909" ind1="C" ind2="O">
    <subfield code="o">oai:fdr.uni-hamburg.de:12671</subfield>
    <subfield code="p">user-csmc</subfield>
    <subfield code="p">user-uhh</subfield>
  </datafield>
  <controlfield tag="001">12671</controlfield>
  <datafield tag="260" ind1=" " ind2=" ">
    <subfield code="c">2023-06-30</subfield>
  </datafield>
  <datafield tag="980" ind1=" " ind2=" ">
    <subfield code="a">dataset</subfield>
  </datafield>
  <datafield tag="024" ind1=" " ind2=" ">
    <subfield code="a">10.25592/uhhfdm.12671</subfield>
    <subfield code="2">doi</subfield>
  </datafield>
  <datafield tag="980" ind1=" " ind2=" ">
    <subfield code="a">user-csmc</subfield>
  </datafield>
  <datafield tag="980" ind1=" " ind2=" ">
    <subfield code="a">user-uhh</subfield>
  </datafield>
  <datafield tag="856" ind1="4" ind2=" ">
    <subfield code="s">42346983</subfield>
    <subfield code="u">https://www.fdr.uni-hamburg.de/record/12671/files/VL_Drawing_Dataset.zip</subfield>
    <subfield code="z">md5:4c79a7d397ab6ad09f14828c75ca2fa9</subfield>
  </datafield>
  <datafield tag="540" ind1=" " ind2=" ">
    <subfield code="u">https://creativecommons.org/licenses/by/4.0/legalcode</subfield>
    <subfield code="a">Creative Commons Attribution 4.0 International</subfield>
  </datafield>
  <datafield tag="773" ind1=" " ind2=" ">
    <subfield code="a">10.25592/uhhfdm.12670</subfield>
    <subfield code="i">isVersionOf</subfield>
    <subfield code="n">doi</subfield>
  </datafield>
  <datafield tag="542" ind1=" " ind2=" ">
    <subfield code="l">open</subfield>
  </datafield>
  <datafield tag="520" ind1=" " ind2=" ">
    <subfield code="a">&lt;p&gt;A minimal dataset of 125 image-text pairs&amp;nbsp;and 10&amp;nbsp;text queries&amp;nbsp;for fine-tuning vision-language models on manuscript images.&amp;nbsp;It is dedicated to the task of text-based image retrieval, and splited into &amp;quot;train&amp;quot; and &amp;quot;test&amp;quot; sets. The train set&amp;nbsp;consists of 100 image-text pairs, while the test set consists of 25 image-text pairs. This dataset is constructed from the following sources:&lt;/p&gt;

&lt;p&gt;- images from&amp;nbsp;the &lt;a href="http://spotting.univ-rouen.fr/"&gt;DocExplore&lt;/a&gt; dataset&amp;nbsp;of medieval manuscripts.&lt;/p&gt;

&lt;p&gt;- images from&amp;nbsp;two manuscripts from Al-Ḥarīrī,&amp;nbsp;&lt;em&gt;Maqāmāt&lt;/em&gt;,&amp;nbsp;&amp;copy; Paris,&amp;nbsp;Biblioth&amp;egrave;que nationale de France. D&amp;eacute;partement des manuscrits,&amp;nbsp;namely MS arabe 3929 and MS arabe 5847.&lt;/p&gt;

&lt;p&gt;- the descriptions in the text files are prepared by&amp;nbsp;Martina Dinelli&lt;/p&gt;

&lt;p&gt;The research for this work was funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) under Germany&amp;#39;s Excellence Strategy &amp;ndash; EXC 2176 &amp;lsquo;Understanding Written Artefacts: Material, Interaction and Transmission in Manuscript Cultures&amp;#39;, project no. 390893796. The research was conducted within the scope of the Centre for the Study of Manuscript Cultures (CSMC) at Universit&amp;auml;t Hamburg.&lt;/p&gt;</subfield>
  </datafield>
  <datafield tag="653" ind1=" " ind2=" ">
    <subfield code="a">Vision-Language Models, Dataset</subfield>
  </datafield>
  <controlfield tag="005">20230630201420.0</controlfield>
  <datafield tag="245" ind1=" " ind2=" ">
    <subfield code="a">Mini-dataset for VL-Models fine-tuning (VL-Tune-dataset-mini)</subfield>
  </datafield>
  <datafield tag="041" ind1=" " ind2=" ">
    <subfield code="a">eng</subfield>
  </datafield>
  <datafield tag="100" ind1=" " ind2=" ">
    <subfield code="a">Hussein Mohammed</subfield>
    <subfield code="u">Universität Hamburg</subfield>
    <subfield code="0">(orcid)0000-0001-5020-3592</subfield>
  </datafield>
  <datafield tag="650" ind1="1" ind2="7">
    <subfield code="a">cc-by</subfield>
    <subfield code="2">opendefinition.org</subfield>
  </datafield>
</record>