Publication detail

Determining Vehicle Turn Counts at Multiple Intersections by Separated Vehicle Classes Using CNNs

FOLENTA, J. ŠPAŇHEL, J. BARTL, V. HEROUT, A.

Original Title

Determining Vehicle Turn Counts at Multiple Intersections by Separated Vehicle Classes Using CNNs

English Title

Determining Vehicle Turn Counts at Multiple Intersections by Separated Vehicle Classes Using CNNs

Type

conference paper

Language

en

Original Abstract

In our submission to the NVIDIA AI City Challenge 2020, we address the problem of counting vehicles by their class at multiple intersections. Our solution is based on counting by tracking principle using convolutional neural networks in detection and tracking steps of the proposed method. We have achieved 6th place on the dataset part A of Track 1 with score S1 Total = 0.8829, (mwRMSE = 4.3616, S1 Effectiveness = 0.9094, S1 Efficiency = 0.8212).

English abstract

In our submission to the NVIDIA AI City Challenge 2020, we address the problem of counting vehicles by their class at multiple intersections. Our solution is based on counting by tracking principle using convolutional neural networks in detection and tracking steps of the proposed method. We have achieved 6th place on the dataset part A of Track 1 with score S1 Total = 0.8829, (mwRMSE = 4.3616, S1 Effectiveness = 0.9094, S1 Efficiency = 0.8212).

Keywords

vehicle counting, vehilce class, intersections, detection, tracking, convolutional neural networks

Released

18.05.2020

Publisher

IEEE Computer Society

Location

Seattle, WA

ISBN

978-1-7281-9360-1

Book

2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)

Edition

IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops

Edition number

NEUVEDEN

Pages from

2544

Pages to

2549

Pages count

6

URL

Documents

BibTex


@inproceedings{BUT168129,
  author="Ján {Folenta} and Jakub {Špaňhel} and Vojtěch {Bartl} and Adam {Herout}",
  title="Determining Vehicle Turn Counts at Multiple Intersections by Separated Vehicle Classes Using CNNs",
  annote="In our submission to the NVIDIA AI City Challenge 2020, we address the problem of
counting vehicles by their class at multiple intersections. Our solution is based
on counting by tracking principle using convolutional neural networks in
detection and tracking steps of the proposed method. We have achieved 6th place
on the dataset part A of Track 1 with score S1 Total = 0.8829, (mwRMSE = 4.3616,
S1 Effectiveness = 0.9094, S1 Efficiency = 0.8212).",
  address="IEEE Computer Society",
  booktitle="2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)",
  chapter="168129",
  doi="10.1109/CVPRW50498.2020.00306",
  edition="IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops",
  howpublished="online",
  institution="IEEE Computer Society",
  number="07",
  year="2020",
  month="may",
  pages="2544--2549",
  publisher="IEEE Computer Society",
  type="conference paper"
}