1 Ocak 2021 Dergi makalesi Açık Erişim

Uner, Onur Can; Aslan, Cem; Ercan, Burak; Ates, Tayfun; Celikcan, Ufuk; Erdem, Aykut; Erdem, Erkut

MARC21 XML

<?xml version='1.0' encoding='UTF-8'?>
<record xmlns="http://www.loc.gov/MARC21/slim">
  <leader>00000nam##2200000uu#4500</leader>
  <datafield tag="700" ind1=" " ind2=" ">
    <subfield code="a">Aslan, Cem</subfield>
    <subfield code="u">Hacettepe Univ, Dept Comp Engn, Ankara, Turkey</subfield>
  </datafield>
  <datafield tag="700" ind1=" " ind2=" ">
    <subfield code="a">Ercan, Burak</subfield>
    <subfield code="u">Hacettepe Univ, Dept Comp Engn, Ankara, Turkey</subfield>
  </datafield>
  <datafield tag="700" ind1=" " ind2=" ">
    <subfield code="a">Ates, Tayfun</subfield>
    <subfield code="u">Hacettepe Univ, Dept Comp Engn, Ankara, Turkey</subfield>
  </datafield>
  <datafield tag="700" ind1=" " ind2=" ">
    <subfield code="a">Celikcan, Ufuk</subfield>
    <subfield code="u">Hacettepe Univ, Dept Comp Engn, Ankara, Turkey</subfield>
  </datafield>
  <datafield tag="700" ind1=" " ind2=" ">
    <subfield code="a">Erdem, Aykut</subfield>
    <subfield code="u">Koc Univ, Dept Comp Engn, Istanbul, Turkey</subfield>
  </datafield>
  <datafield tag="700" ind1=" " ind2=" ">
    <subfield code="a">Erdem, Erkut</subfield>
    <subfield code="u">Hacettepe Univ, Dept Comp Engn, Ankara, Turkey</subfield>
  </datafield>
  <datafield tag="909" ind1="C" ind2="4">
    <subfield code="p">SIGNAL PROCESSING-IMAGE COMMUNICATION</subfield>
    <subfield code="v">97</subfield>
  </datafield>
  <datafield tag="980" ind1=" " ind2=" ">
    <subfield code="a">user-tubitak-destekli-proje-yayinlari</subfield>
  </datafield>
  <datafield tag="540" ind1=" " ind2=" ">
    <subfield code="a">Creative Commons Attribution</subfield>
    <subfield code="u">http://www.opendefinition.org/licenses/cc-by</subfield>
  </datafield>
  <datafield tag="024" ind1=" " ind2=" ">
    <subfield code="a">10.1016/j.image.2021.116335</subfield>
    <subfield code="2">doi</subfield>
  </datafield>
  <datafield tag="245" ind1=" " ind2=" ">
    <subfield code="a">Synthetic18K: Learning better representations for person re-ID and attribute recognition from 1.4 million synthetic images</subfield>
  </datafield>
  <datafield tag="100" ind1=" " ind2=" ">
    <subfield code="a">Uner, Onur Can</subfield>
    <subfield code="u">Bilkent Univ, Dept Comp Engn, Ankara, Turkey</subfield>
  </datafield>
  <datafield tag="909" ind1="C" ind2="O">
    <subfield code="o">oai:aperta.ulakbim.gov.tr:236126</subfield>
    <subfield code="p">user-tubitak-destekli-proje-yayinlari</subfield>
  </datafield>
  <datafield tag="650" ind1="1" ind2="7">
    <subfield code="2">opendefinition.org</subfield>
    <subfield code="a">cc-by</subfield>
  </datafield>
  <datafield tag="260" ind1=" " ind2=" ">
    <subfield code="c">2021-01-01</subfield>
  </datafield>
  <datafield tag="856" ind1="4" ind2=" ">
    <subfield code="u">https://aperta.ulakbim.gov.trrecord/236126/files/bib-0d5cae8f-ca33-4585-abe3-e0c0704b9878.txt</subfield>
    <subfield code="z">md5:ae5a7776b608892c290f47c15a7a13ed</subfield>
    <subfield code="s">258</subfield>
  </datafield>
  <datafield tag="542" ind1=" " ind2=" ">
    <subfield code="l">open</subfield>
  </datafield>
  <controlfield tag="005">20221007093516.0</controlfield>
  <controlfield tag="001">236126</controlfield>
  <datafield tag="980" ind1=" " ind2=" ">
    <subfield code="a">publication</subfield>
    <subfield code="b">article</subfield>
  </datafield>
  <datafield tag="520" ind1=" " ind2=" ">
    <subfield code="a">Learning robust representations is critical for the success of person re-identification and attribute recognition systems. However, to achieve this, we must use a large dataset of diverse person images as well as annotations of identity labels and/or a set of different attributes. Apart from the obvious concerns about privacy issues, the manual annotation process is both time consuming and too costly. In this paper, we instead propose to use synthetic person images for addressing these difficulties. Specifically, we first introduce Synthetic18K, a large-scale dataset of over 1 million computer generated person images of 18K unique identities with relevant attributes. Moreover, we demonstrate that pretraining of simple deep architectures on Synthetic18K for person re-identification and attribute recognition and then fine-tuning on real data leads to significant improvements in prediction performances, giving results better than or comparable to state-of-the-art models.</subfield>
  </datafield>
</record>