Robust vehicle tracking for urban traffic videos at intersections

C. Li, A. Chiang, G. Dobler, Y. Wang, K. Xie, K. Ozbay, M. Ghandehari, J. Zhou, D. Wang

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

We develop a robust, unsupervised vehicle tracking system for videos of very congested road intersections in urban environments. Raw tracklets from the standard Kanade-Lucas-Tomasi tracking algorithm are treated as sample points and grouped to form different vehicle candidates. Each tracklet is described by multiple features including position, velocity, and a foreground score derived from robust PCA background subtraction. By considering each tracklet as a node in a graph, we build the adjacency matrix for the graph based on the feature similarity between the tracklets and group these tracklets using spectral embedding and Dirichelet Process Gaussian Mixture Models. The proposed system yields excellent performance for traffic videos captured in urban environments and highways.

Original languageEnglish (US)
Title of host publication2016 13th IEEE International Conference on Advanced Video and Signal Based Surveillance, AVSS 2016
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages207-213
Number of pages7
ISBN (Electronic)9781509038114
DOIs
StatePublished - Nov 7 2016
Event13th IEEE International Conference on Advanced Video and Signal Based Surveillance, AVSS 2016 - Colorado Springs, United States
Duration: Aug 23 2016Aug 26 2016

Other

Other13th IEEE International Conference on Advanced Video and Signal Based Surveillance, AVSS 2016
CountryUnited States
CityColorado Springs
Period8/23/168/26/16

Fingerprint

Telecommunication traffic
intersections
traffic
vehicles
roads
subtraction
embedding
matrices

ASJC Scopus subject areas

  • Signal Processing
  • Media Technology
  • Instrumentation
  • Computer Vision and Pattern Recognition

Cite this

Li, C., Chiang, A., Dobler, G., Wang, Y., Xie, K., Ozbay, K., ... Wang, D. (2016). Robust vehicle tracking for urban traffic videos at intersections. In 2016 13th IEEE International Conference on Advanced Video and Signal Based Surveillance, AVSS 2016 (pp. 207-213). [7738075] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/AVSS.2016.7738075

Robust vehicle tracking for urban traffic videos at intersections. / Li, C.; Chiang, A.; Dobler, G.; Wang, Y.; Xie, K.; Ozbay, K.; Ghandehari, M.; Zhou, J.; Wang, D.

2016 13th IEEE International Conference on Advanced Video and Signal Based Surveillance, AVSS 2016. Institute of Electrical and Electronics Engineers Inc., 2016. p. 207-213 7738075.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Li, C, Chiang, A, Dobler, G, Wang, Y, Xie, K, Ozbay, K, Ghandehari, M, Zhou, J & Wang, D 2016, Robust vehicle tracking for urban traffic videos at intersections. in 2016 13th IEEE International Conference on Advanced Video and Signal Based Surveillance, AVSS 2016., 7738075, Institute of Electrical and Electronics Engineers Inc., pp. 207-213, 13th IEEE International Conference on Advanced Video and Signal Based Surveillance, AVSS 2016, Colorado Springs, United States, 8/23/16. https://doi.org/10.1109/AVSS.2016.7738075
Li C, Chiang A, Dobler G, Wang Y, Xie K, Ozbay K et al. Robust vehicle tracking for urban traffic videos at intersections. In 2016 13th IEEE International Conference on Advanced Video and Signal Based Surveillance, AVSS 2016. Institute of Electrical and Electronics Engineers Inc. 2016. p. 207-213. 7738075 https://doi.org/10.1109/AVSS.2016.7738075
Li, C. ; Chiang, A. ; Dobler, G. ; Wang, Y. ; Xie, K. ; Ozbay, K. ; Ghandehari, M. ; Zhou, J. ; Wang, D. / Robust vehicle tracking for urban traffic videos at intersections. 2016 13th IEEE International Conference on Advanced Video and Signal Based Surveillance, AVSS 2016. Institute of Electrical and Electronics Engineers Inc., 2016. pp. 207-213
@inproceedings{8b08ce34e2b14f53b9be49fd66d6bb10,
title = "Robust vehicle tracking for urban traffic videos at intersections",
abstract = "We develop a robust, unsupervised vehicle tracking system for videos of very congested road intersections in urban environments. Raw tracklets from the standard Kanade-Lucas-Tomasi tracking algorithm are treated as sample points and grouped to form different vehicle candidates. Each tracklet is described by multiple features including position, velocity, and a foreground score derived from robust PCA background subtraction. By considering each tracklet as a node in a graph, we build the adjacency matrix for the graph based on the feature similarity between the tracklets and group these tracklets using spectral embedding and Dirichelet Process Gaussian Mixture Models. The proposed system yields excellent performance for traffic videos captured in urban environments and highways.",
author = "C. Li and A. Chiang and G. Dobler and Y. Wang and K. Xie and K. Ozbay and M. Ghandehari and J. Zhou and D. Wang",
year = "2016",
month = "11",
day = "7",
doi = "10.1109/AVSS.2016.7738075",
language = "English (US)",
pages = "207--213",
booktitle = "2016 13th IEEE International Conference on Advanced Video and Signal Based Surveillance, AVSS 2016",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
address = "United States",

}

TY - GEN

T1 - Robust vehicle tracking for urban traffic videos at intersections

AU - Li, C.

AU - Chiang, A.

AU - Dobler, G.

AU - Wang, Y.

AU - Xie, K.

AU - Ozbay, K.

AU - Ghandehari, M.

AU - Zhou, J.

AU - Wang, D.

PY - 2016/11/7

Y1 - 2016/11/7

N2 - We develop a robust, unsupervised vehicle tracking system for videos of very congested road intersections in urban environments. Raw tracklets from the standard Kanade-Lucas-Tomasi tracking algorithm are treated as sample points and grouped to form different vehicle candidates. Each tracklet is described by multiple features including position, velocity, and a foreground score derived from robust PCA background subtraction. By considering each tracklet as a node in a graph, we build the adjacency matrix for the graph based on the feature similarity between the tracklets and group these tracklets using spectral embedding and Dirichelet Process Gaussian Mixture Models. The proposed system yields excellent performance for traffic videos captured in urban environments and highways.

AB - We develop a robust, unsupervised vehicle tracking system for videos of very congested road intersections in urban environments. Raw tracklets from the standard Kanade-Lucas-Tomasi tracking algorithm are treated as sample points and grouped to form different vehicle candidates. Each tracklet is described by multiple features including position, velocity, and a foreground score derived from robust PCA background subtraction. By considering each tracklet as a node in a graph, we build the adjacency matrix for the graph based on the feature similarity between the tracklets and group these tracklets using spectral embedding and Dirichelet Process Gaussian Mixture Models. The proposed system yields excellent performance for traffic videos captured in urban environments and highways.

UR - http://www.scopus.com/inward/record.url?scp=85003827440&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85003827440&partnerID=8YFLogxK

U2 - 10.1109/AVSS.2016.7738075

DO - 10.1109/AVSS.2016.7738075

M3 - Conference contribution

SP - 207

EP - 213

BT - 2016 13th IEEE International Conference on Advanced Video and Signal Based Surveillance, AVSS 2016

PB - Institute of Electrical and Electronics Engineers Inc.

ER -