Publication detail

Vehicle Re-Identification and Multi-Camera Tracking in Challenging City-Scale Environment

ŠPAŇHEL, J. BARTL, V. JURÁNEK, R. HEROUT, A.

Original Title

Vehicle Re-Identification and Multi-Camera Tracking in Challenging City-Scale Environment

English Title

Vehicle Re-Identification and Multi-Camera Tracking in Challenging City-Scale Environment

Type

conference paper

Language

en

Original Abstract

In our submission to the NVIDIA AI City Challenge, we address vehicle re-identification and vehicle multi-camera tracking. Our approach to vehicle re-identification is based on the extraction of visual features and aggregation of these features in the temporal domain to obtain a single feature descriptor for the whole observed track. For multi-camera tracking, we proposed a method for matching vehicles by the position of trajectory points in real-world space (linear coordinate system). Furthermore, we use CNN for the vehicle re-identification task to filter out false matches generated by proposed positional matching method for better results.

English abstract

In our submission to the NVIDIA AI City Challenge, we address vehicle re-identification and vehicle multi-camera tracking. Our approach to vehicle re-identification is based on the extraction of visual features and aggregation of these features in the temporal domain to obtain a single feature descriptor for the whole observed track. For multi-camera tracking, we proposed a method for matching vehicles by the position of trajectory points in real-world space (linear coordinate system). Furthermore, we use CNN for the vehicle re-identification task to filter out false matches generated by proposed positional matching method for better results.

Keywords

vehicle re-identification, vehicle multi-camera tracking, city-scale environment, camera calibration, neural networks, nvidia ai city challenge

Released

03.07.2019

Publisher

IEEE Computer Society

Location

Long Beach

ISBN

2160-7516

Year of study

2019

Number

1

Pages from

150

Pages to

158

Pages count

9

URL

http://openaccess.thecvf.com/content_CVPRW_2019/html/AI_City/Spanhel_Vehicle_Re-Identifiation_and_Multi-Camera_Tracking_in_Challenging_City-Scale_Environment_CVPRW_2019_paper.html

Documents

BibTex 


@inproceedings{BUT162081,
  author="Jakub {Špaňhel} and Vojtěch {Bartl} and Roman {Juránek} and Adam {Herout}",
  title="Vehicle Re-Identification and Multi-Camera Tracking in Challenging City-Scale Environment",
  annote="In our submission to the NVIDIA AI City Challenge, we address vehicle
re-identification and vehicle multi-camera tracking. Our approach to vehicle
re-identification is based on the extraction of visual features and aggregation
of these features in the temporal domain to obtain a single feature descriptor
for the whole observed track. For multi-camera tracking, we proposed a method for
matching vehicles by the position of trajectory points in real-world space
(linear coordinate system). Furthermore, we use CNN for the vehicle
re-identification task to filter out false matches generated by proposed
positional matching method for better results.",
  address="IEEE Computer Society",
  booktitle="2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)",
  chapter="162081",
  edition="IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops",
  howpublished="online",
  institution="IEEE Computer Society",
  number="1",
  year="2019",
  month="july",
  pages="150--158",
  publisher="IEEE Computer Society",
  type="conference paper"
}