Publication Details

Density-Based Vehicle Counting with Unsupervised Scale Selection

DOBEŠ Petr, ŠPAŇHEL Jakub, BARTL Vojtěch, JURÁNEK Roman and HEROUT Adam. Density-Based Vehicle Counting with Unsupervised Scale Selection. In: Digital Image Computing: Techniques and Applications 2020. Melbourne: Institute of Electrical and Electronics Engineers, 2020, pp. 1-8. ISBN 978-1-7281-9108-9. Available from: http://www.dicta2020.org/wp-content/uploads/2020/09/22_CameraReady.pdf
Czech title
Odhad počtu vozidel v obraze pomocí pravděpodobnostní mapy s automatickým výběrem měřítka
Type
conference paper
Language
english
Authors
URL
Keywords

Deep learning; Density estimation; Object counting; Traffic surveillance; Vehicle counting

Abstract

A significant hurdle within any counting task is the variance in scale of the objects to be counted. While size changes of some extent can be induced by perspective distortion, more severe scale differences can easily occur, e.g. in case of images taken by a drone from different elevations above the ground. The aim of our work is to overcome this issue by leveraging only lightweight dot annotations and a minimum level of training supervision. We propose a modification to the Stacked Hourglass network which enables the model to process multiple input scales and to automatically select the most suitable candidate using a quality score. We alter the training procedure to enable learning of the quality scores while avoiding their direct supervision, and thus without requiring any additional annotation effort. We evaluate our method on three standard datasets: PUCPR+, TRANCOS and CARPK. The obtained results are on par with current state-of-the-art methods while being more robust towards significant variations in input scale.

Published
2020
Pages
1-8
Proceedings
Digital Image Computing: Techniques and Applications 2020
Conference
Digital Image Computing: Techniques and Applications 2020, Melbourne, AU
ISBN
978-1-7281-9108-9
Publisher
Institute of Electrical and Electronics Engineers
Place
Melbourne, AU
DOI
EID Scopus
BibTeX
@INPROCEEDINGS{FITPUB12360,
   author = "Petr Dobe\v{s} and Jakub \v{S}pa\v{n}hel and Vojt\v{e}ch Bartl and Roman Jur\'{a}nek and Adam Herout",
   title = "Density-Based Vehicle Counting with Unsupervised Scale Selection",
   pages = "1--8",
   booktitle = "Digital Image Computing: Techniques and Applications 2020",
   year = 2020,
   location = "Melbourne, AU",
   publisher = "Institute of Electrical and Electronics Engineers",
   ISBN = "978-1-7281-9108-9",
   doi = "10.1109/DICTA51227.2020.9363401",
   language = "english",
   url = "https://www.fit.vut.cz/research/publication/12360"
}
Back to top