Conference paper Open Access

Through-Screen Visible Light Sensing Empowered by Embedded Deep Learning

Liu, Hao; Ye, Hanting; Yang, Jie; Wang, Qing


MARC21 XML Export

<?xml version='1.0' encoding='UTF-8'?>
<record xmlns="http://www.loc.gov/MARC21/slim">
  <leader>00000nam##2200000uu#4500</leader>
  <controlfield tag="005">20211105134832.0</controlfield>
  <controlfield tag="001">5646942</controlfield>
  <datafield tag="711" ind1=" " ind2=" ">
    <subfield code="d">15-17 November 2021</subfield>
    <subfield code="g">AIChallengeIoT</subfield>
    <subfield code="a">Workshop on Workshop on Challenges in Artificial Intelligence and Machine Learning for Internet of Things in conjuction with ACM SenSys 2021</subfield>
    <subfield code="c">Coimbra, Portugal</subfield>
  </datafield>
  <datafield tag="700" ind1=" " ind2=" ">
    <subfield code="u">TU Delft</subfield>
    <subfield code="a">Ye, Hanting</subfield>
  </datafield>
  <datafield tag="700" ind1=" " ind2=" ">
    <subfield code="u">TU Delft</subfield>
    <subfield code="a">Yang, Jie</subfield>
  </datafield>
  <datafield tag="700" ind1=" " ind2=" ">
    <subfield code="u">TU Delft</subfield>
    <subfield code="a">Wang, Qing</subfield>
  </datafield>
  <datafield tag="856" ind1="4" ind2=" ">
    <subfield code="s">1297786</subfield>
    <subfield code="z">md5:9fc03485ef13a9329b8301fc86ffa802</subfield>
    <subfield code="u">https://zenodo.org/record/5646942/files/AIChallengeIoT2021_Through-Screen VLS.pdf</subfield>
  </datafield>
  <datafield tag="542" ind1=" " ind2=" ">
    <subfield code="l">open</subfield>
  </datafield>
  <datafield tag="856" ind1="4" ind2=" ">
    <subfield code="y">Conference website</subfield>
    <subfield code="u">https://aichallengeiot.github.io/</subfield>
  </datafield>
  <datafield tag="260" ind1=" " ind2=" ">
    <subfield code="c">2021-11-05</subfield>
  </datafield>
  <datafield tag="909" ind1="C" ind2="O">
    <subfield code="p">openaire</subfield>
    <subfield code="o">oai:zenodo.org:5646942</subfield>
  </datafield>
  <datafield tag="100" ind1=" " ind2=" ">
    <subfield code="u">TU Delft</subfield>
    <subfield code="a">Liu, Hao</subfield>
  </datafield>
  <datafield tag="245" ind1=" " ind2=" ">
    <subfield code="a">Through-Screen Visible Light Sensing Empowered by Embedded Deep Learning</subfield>
  </datafield>
  <datafield tag="536" ind1=" " ind2=" ">
    <subfield code="c">814215</subfield>
    <subfield code="a">European Training Network in Low-energy Visible Light IoT Systems</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="650" ind1="1" ind2="7">
    <subfield code="a">cc-by</subfield>
    <subfield code="2">opendefinition.org</subfield>
  </datafield>
  <datafield tag="520" ind1=" " ind2=" ">
    <subfield code="a">&lt;p&gt;Motivated by the trend of realizing full screens on devices such as smartphones, in this work we propose through-screen sensing with visible light for the application of fingertip air-writing. The system can recognize handwritten digits with under-screen photodiodes as the receiver. The key idea is to recognize the weak light reflected by the finger when the finger writes the digits on top of a screen. The proposed air-writing system has immunity to scene changes because it has a fixed screen light source. However, the screen is a double-edged sword as both a signal source and a noise source. We propose a data preprocessing method to reduce the interference of the screen as a noise source. We design an embedded deep learning model, a customized model ConvRNN, to model the spatial and temporal patterns in the dynamic and weak reflected signal for air-writing digits recognition. The evaluation results show that our through-screen fingertip air-writing system with visible light can achieve accuracy up to 91%. Results further show that the size of the customized ConvRNN model can be reduced by 94% with less&lt;br&gt;
than a 10% drop in performance.&lt;/p&gt;</subfield>
  </datafield>
  <datafield tag="024" ind1=" " ind2=" ">
    <subfield code="a">10.1145/3485730.3493454</subfield>
    <subfield code="2">doi</subfield>
  </datafield>
  <datafield tag="980" ind1=" " ind2=" ">
    <subfield code="a">publication</subfield>
    <subfield code="b">conferencepaper</subfield>
  </datafield>
</record>
22
11
views
downloads
Views 22
Downloads 11
Data volume 14.3 MB
Unique views 22
Unique downloads 11

Share

Cite as