A Distributed Tracking Algorithm for Counting People in Video
- Authors: Kuplyakov D.A.1,2, Shalnov E.V.1, Konushin V.S.2, Konushin A.S.1
-
Affiliations:
- Faculty of Computational Mathematics and Cybernetics, Moscow State University
- Video Analysis Technologies
- Issue: Vol 45, No 4 (2019)
- Pages: 163-170
- Section: Article
- URL: https://journals.rcsi.science/0361-7688/article/view/176816
- DOI: https://doi.org/10.1134/S0361768819040042
- ID: 176816
Cite item
Abstract
The problem of tracking people in a video stream with the aim of counting them is studied. Modern video surveillance systems, such as the Moscow video surveillance system, use hundreds of thousands of cameras. The use of modern methods developed for working on a single computer with an expensive graphical processor is economically inefficient for such large-scale systems. In this paper, a distributed tracking algorithm is proposed. It makes it possible to reduce the amount of computational resources due to detecting people in a sparse set of frames. The detection is performed on servers installed in a data center, while the video stream is processed by local camera computation nodes. The experimental evaluation showed that the proposed algorithm provides acceptable quality at the detection rate of 4/3 Hz.
About the authors
D. A. Kuplyakov
Faculty of Computational Mathematics and Cybernetics, Moscow State University; Video Analysis Technologies
Author for correspondence.
Email: denis.kuplyakov@graphics.cs.msu.ru
Russian Federation, Moscow, 119899; Moscow, 119634
E. V. Shalnov
Faculty of Computational Mathematics and Cybernetics, Moscow State University
Author for correspondence.
Email: eshalnov@graphics.cs.msu.ru
Russian Federation, Moscow, 119899
V. S. Konushin
Video Analysis Technologies
Author for correspondence.
Email: vadim@tevian.ru
Russian Federation, Moscow, 119634
A. S. Konushin
Faculty of Computational Mathematics and Cybernetics, Moscow State University
Author for correspondence.
Email: ktosh@graphics.cs.msu.ru
Russian Federation, Moscow, 119899