ML Anthology

Multi-Class Multi-Movement Vehicle Counting Based on CenterTrack

Viktor Kocur, Milan Ftácnik
CVPRW 2021 pp. 4009-4015
doi:10.1109/CVPRW53098.2021.00452 /cvprw/2021/kocur2021cvprw-multiclass/

Abstract

In this paper we present our approach to the Track 1 of the 2021 AI City Challenge. The goal of the challenge track is to to analyse footage captured with traffic cameras by counting the number of vehicles performing various predefined motions of interest. Our approach is based on the CenterTrack object detection and tracking neural network used in conjunction with a simple IoU-based tracking algorithm. In the public evaluation server our system achieved the S1 score of 0.8449 placing it at the 8th place on the public leaderboard.

CVPRW Semantic Scholar

Cite

Text

Kocur and Ftácnik. "Multi-Class Multi-Movement Vehicle Counting Based on CenterTrack." IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2021. doi:10.1109/CVPRW53098.2021.00452

Markdown

[Kocur and Ftácnik. "Multi-Class Multi-Movement Vehicle Counting Based on CenterTrack." IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2021.](https://mlanthology.org/cvprw/2021/kocur2021cvprw-multiclass/) doi:10.1109/CVPRW53098.2021.00452

BibTeX

@inproceedings{kocur2021cvprw-multiclass,
  title     = {{Multi-Class Multi-Movement Vehicle Counting Based on CenterTrack}},
  author    = {Kocur, Viktor and Ftácnik, Milan},
  booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops},
  year      = {2021},
  pages     = {4009-4015},
  doi       = {10.1109/CVPRW53098.2021.00452},
  url       = {https://mlanthology.org/cvprw/2021/kocur2021cvprw-multiclass/}
}