ML Anthology

Graph@FIT Submission to the NVIDIA AI City Challenge 2018

Jakub Sochor, Jakub Spanhel, Roman Juránek, Petr Dobes, Adam Herout
CVPRW 2018 pp. 77-84
doi:10.1109/CVPRW.2018.00018 /cvprw/2018/sochor2018cvprw-graph/

Abstract

In our submission to the NVIDIA AI City Challenge, we address speed measurement of vehicles and vehicle re-identification. For both these tasks, we use a calibration method based on extracted vanishing points. We detect and track vehicles by a CNN-based detector and we construct 3D bounding boxes for all vehicles. For the speed measurement task, we estimate the speed from the movement of the bounding box in the 3D space using the calibration. Our approach to vehicle re-identification is based on extraction of visual features from "unpacked" images of the vehicles. The features are aggregated in temporal domain to obtain a single feature descriptor for the whole track. Furthermore, we utilize a validation network to improve the re-identification accuracy.

PDF CVPRW Semantic Scholar

Cite

Text

Sochor et al. "Graph@FIT Submission to the NVIDIA AI City Challenge 2018." IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2018. doi:10.1109/CVPRW.2018.00018

Markdown

[Sochor et al. "Graph@FIT Submission to the NVIDIA AI City Challenge 2018." IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2018.](https://mlanthology.org/cvprw/2018/sochor2018cvprw-graph/) doi:10.1109/CVPRW.2018.00018

BibTeX

@inproceedings{sochor2018cvprw-graph,
  title     = {{Graph@FIT Submission to the NVIDIA AI City Challenge 2018}},
  author    = {Sochor, Jakub and Spanhel, Jakub and Juránek, Roman and Dobes, Petr and Herout, Adam},
  booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops},
  year      = {2018},
  pages     = {77-84},
  doi       = {10.1109/CVPRW.2018.00018},
  url       = {https://mlanthology.org/cvprw/2018/sochor2018cvprw-graph/}
}