Result Details
Graph@FIT Submission to the NVIDIA AI City Challenge 2018
Špaňhel Jakub, Ing., Ph.D., DCGM (FIT)
Juránek Roman, Ing., Ph.D., DCGM (FIT)
Dobeš Petr, Ing., DCGM (FIT)
Herout Adam, prof. Ing., Ph.D., DCGM (FIT)
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.
vehicle speed measurement, vehicle re-identification, challenge, camera calibration
@inproceedings{BUT155029,
author="Jakub {Sochor} and Jakub {Špaňhel} and Roman {Juránek} and Petr {Dobeš} and Adam {Herout}",
title="Graph@FIT Submission to the NVIDIA AI City Challenge 2018",
booktitle="NVIDIA AI City Challenge 2018 (CVPRW)",
year="2018",
pages="77--84",
publisher="IEEE Computer Society",
address="Salt Lake City",
doi="10.1109/CVPRW.2018.00018",
isbn="978-1-5386-6100-0",
url="https://ieeexplore.ieee.org/document/8575345"
}SMARTCarPark - Surveillance Monitoring, Analysis and Re-identification of Traffic for Enhanced Car Parking, TAČR, Program na podporu aplikovaného výzkumu a experimentálního vývoje EPSILON (2015-2025), TH03010529, start: 2018-01-01, end: 2020-12-31, completed