A Temporal fusion approach for video classifcation with convolutional and LSTM neural networks applied to violence detection

<?xml version="1.0" encoding="UTF-8"?><collection xmlns="http://www.loc.gov/MARC21/slim" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.loc.gov/MARC21/slim http://www.loc.gov/standards/marcxml/schema/MARC21slim.xsd">
  <record>
    <leader>00000cab a2200000   4500</leader>
    <controlfield tag="001">MAP20210011740</controlfield>
    <controlfield tag="003">MAP</controlfield>
    <controlfield tag="005">20220911190022.0</controlfield>
    <controlfield tag="008">210413e20210215esp|||p      |0|||b|spa d</controlfield>
    <datafield tag="040" ind1=" " ind2=" ">
      <subfield code="a">MAP</subfield>
      <subfield code="b">spa</subfield>
      <subfield code="d">MAP</subfield>
    </datafield>
    <datafield tag="084" ind1=" " ind2=" ">
      <subfield code="a">922.134</subfield>
    </datafield>
    <datafield tag="100" ind1=" " ind2=" ">
      <subfield code="0">MAPA20200022954</subfield>
      <subfield code="a">Oliveira Lima, Jean Phelipe de </subfield>
    </datafield>
    <datafield tag="245" ind1="1" ind2="2">
      <subfield code="a">A Temporal fusion approach for video classifcation with convolutional and LSTM neural networks applied to violence detection</subfield>
      <subfield code="c">Jean Phelipe de Oliveira Lima, Carlos Maurício Seródio Figueiredo</subfield>
    </datafield>
    <datafield tag="520" ind1=" " ind2=" ">
      <subfield code="a">In modern smart cities, there is a quest for the highest level of integration and automation service. In the surveillance sector, one of the main challenges is to automate the analysis of videos in real-time to identify critical situations. This paper presents intelligent models based on Convolutional Neural Networks (in which the MobileNet, InceptionV3 and VGG16 networks had used), LSTM networks and feedforward networks for the task of classifying videos under the classes "Violence" and "Non-Violence", using for this the RLVS database. Di rent data representations held used according to the Temporal Fusion techniques. The best outcome achieved was 0.91 and 0.90 of Accuracy and F1-Score, respectively, a higher result compared to those found in similar researches for works conducted on the same database. </subfield>
    </datafield>
    <datafield tag="650" ind1=" " ind2="4">
      <subfield code="0">MAPA20080611200</subfield>
      <subfield code="a">Inteligencia artificial</subfield>
    </datafield>
    <datafield tag="650" ind1=" " ind2="4">
      <subfield code="0">MAPA20080568009</subfield>
      <subfield code="a">Automatización</subfield>
    </datafield>
    <datafield tag="650" ind1=" " ind2="4">
      <subfield code="0">MAPA20080552954</subfield>
      <subfield code="a">Violencia</subfield>
    </datafield>
    <datafield tag="700" ind1="1" ind2=" ">
      <subfield code="0">MAPA20200023012</subfield>
      <subfield code="a">Seródio Figueiredo, Carlos Maurício </subfield>
    </datafield>
    <datafield tag="773" ind1="0" ind2=" ">
      <subfield code="w">MAP20200034445</subfield>
      <subfield code="t">Revista Iberoamericana de Inteligencia Artificial</subfield>
      <subfield code="d">IBERAMIA, Sociedad Iberoamericana de Inteligencia Artificial , 2018-</subfield>
      <subfield code="x">1988-3064</subfield>
      <subfield code="g">15/02/2021 Volumen 24 Número 67 - febrero 2021 , p. 40-50</subfield>
    </datafield>
    <datafield tag="856" ind1=" " ind2=" ">
      <subfield code="q">application/pdf</subfield>
      <subfield code="w">1110631</subfield>
      <subfield code="y">Recurso electrónico / Electronic resource</subfield>
    </datafield>
  </record>
</collection>