Online pedestrian tracking with multi-stage re-identification

Yi Fan Jiang, Hyunhak Shin, Jaeyong Ju, Hanseok Ko

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

1 Citation (Scopus)

Abstract

Nowadays the task of tracking pedestrians is often addressed within a tracking-by-detection framework, which in most cases entails that the position of each target has been detected before tracking begins. However in some cases, a pedestrian who is being tracked may be obscured by other targets or obstacles, and during this period they may change their trajectory or speed (track drift), and sometimes such a target may leave the FOV (Field of View) [10] but appear again later. These temporary disappearances and absence of detections disrupt the work of the detectors to such an extent that there is a significant decline in performance. In this paper, we propose a novel approach to pedestrian tracking based on multi-stage re-identification. To deal with the problems discussed above, the proposed framework is comprised of a two-stage re-identification algorithm dealing with cases of track drift and re-entry into the FOV individually, in order to match the identities of lost and reappeared targets through a comparison of the affinities between their appearance, size and position, and also to update the status of re-identified targets through this assessment. The experimental results demonstrate that this framework can effectively handle complex temporary lost and re-entry situations with robustness, and that its performance is state of the art.

Original languageEnglish
Title of host publication2017 14th IEEE International Conference on Advanced Video and Signal Based Surveillance, AVSS 2017
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781538629390
DOIs
Publication statusPublished - 2017 Oct 20
Event14th IEEE International Conference on Advanced Video and Signal Based Surveillance, AVSS 2017 - Lecce, Italy
Duration: 2017 Aug 292017 Sep 1

Other

Other14th IEEE International Conference on Advanced Video and Signal Based Surveillance, AVSS 2017
CountryItaly
CityLecce
Period17/8/2917/9/1

Fingerprint

Reentry
Trajectories
Detectors

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition
  • Signal Processing

Cite this

Jiang, Y. F., Shin, H., Ju, J., & Ko, H. (2017). Online pedestrian tracking with multi-stage re-identification. In 2017 14th IEEE International Conference on Advanced Video and Signal Based Surveillance, AVSS 2017 [8078505] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/AVSS.2017.8078505

Online pedestrian tracking with multi-stage re-identification. / Jiang, Yi Fan; Shin, Hyunhak; Ju, Jaeyong; Ko, Hanseok.

2017 14th IEEE International Conference on Advanced Video and Signal Based Surveillance, AVSS 2017. Institute of Electrical and Electronics Engineers Inc., 2017. 8078505.

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

Jiang, YF, Shin, H, Ju, J & Ko, H 2017, Online pedestrian tracking with multi-stage re-identification. in 2017 14th IEEE International Conference on Advanced Video and Signal Based Surveillance, AVSS 2017., 8078505, Institute of Electrical and Electronics Engineers Inc., 14th IEEE International Conference on Advanced Video and Signal Based Surveillance, AVSS 2017, Lecce, Italy, 17/8/29. https://doi.org/10.1109/AVSS.2017.8078505
Jiang YF, Shin H, Ju J, Ko H. Online pedestrian tracking with multi-stage re-identification. In 2017 14th IEEE International Conference on Advanced Video and Signal Based Surveillance, AVSS 2017. Institute of Electrical and Electronics Engineers Inc. 2017. 8078505 https://doi.org/10.1109/AVSS.2017.8078505
Jiang, Yi Fan ; Shin, Hyunhak ; Ju, Jaeyong ; Ko, Hanseok. / Online pedestrian tracking with multi-stage re-identification. 2017 14th IEEE International Conference on Advanced Video and Signal Based Surveillance, AVSS 2017. Institute of Electrical and Electronics Engineers Inc., 2017.
@inproceedings{f43902806b754471a5915fd7152fde32,
title = "Online pedestrian tracking with multi-stage re-identification",
abstract = "Nowadays the task of tracking pedestrians is often addressed within a tracking-by-detection framework, which in most cases entails that the position of each target has been detected before tracking begins. However in some cases, a pedestrian who is being tracked may be obscured by other targets or obstacles, and during this period they may change their trajectory or speed (track drift), and sometimes such a target may leave the FOV (Field of View) [10] but appear again later. These temporary disappearances and absence of detections disrupt the work of the detectors to such an extent that there is a significant decline in performance. In this paper, we propose a novel approach to pedestrian tracking based on multi-stage re-identification. To deal with the problems discussed above, the proposed framework is comprised of a two-stage re-identification algorithm dealing with cases of track drift and re-entry into the FOV individually, in order to match the identities of lost and reappeared targets through a comparison of the affinities between their appearance, size and position, and also to update the status of re-identified targets through this assessment. The experimental results demonstrate that this framework can effectively handle complex temporary lost and re-entry situations with robustness, and that its performance is state of the art.",
author = "Jiang, {Yi Fan} and Hyunhak Shin and Jaeyong Ju and Hanseok Ko",
year = "2017",
month = "10",
day = "20",
doi = "10.1109/AVSS.2017.8078505",
language = "English",
booktitle = "2017 14th IEEE International Conference on Advanced Video and Signal Based Surveillance, AVSS 2017",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - Online pedestrian tracking with multi-stage re-identification

AU - Jiang, Yi Fan

AU - Shin, Hyunhak

AU - Ju, Jaeyong

AU - Ko, Hanseok

PY - 2017/10/20

Y1 - 2017/10/20

N2 - Nowadays the task of tracking pedestrians is often addressed within a tracking-by-detection framework, which in most cases entails that the position of each target has been detected before tracking begins. However in some cases, a pedestrian who is being tracked may be obscured by other targets or obstacles, and during this period they may change their trajectory or speed (track drift), and sometimes such a target may leave the FOV (Field of View) [10] but appear again later. These temporary disappearances and absence of detections disrupt the work of the detectors to such an extent that there is a significant decline in performance. In this paper, we propose a novel approach to pedestrian tracking based on multi-stage re-identification. To deal with the problems discussed above, the proposed framework is comprised of a two-stage re-identification algorithm dealing with cases of track drift and re-entry into the FOV individually, in order to match the identities of lost and reappeared targets through a comparison of the affinities between their appearance, size and position, and also to update the status of re-identified targets through this assessment. The experimental results demonstrate that this framework can effectively handle complex temporary lost and re-entry situations with robustness, and that its performance is state of the art.

AB - Nowadays the task of tracking pedestrians is often addressed within a tracking-by-detection framework, which in most cases entails that the position of each target has been detected before tracking begins. However in some cases, a pedestrian who is being tracked may be obscured by other targets or obstacles, and during this period they may change their trajectory or speed (track drift), and sometimes such a target may leave the FOV (Field of View) [10] but appear again later. These temporary disappearances and absence of detections disrupt the work of the detectors to such an extent that there is a significant decline in performance. In this paper, we propose a novel approach to pedestrian tracking based on multi-stage re-identification. To deal with the problems discussed above, the proposed framework is comprised of a two-stage re-identification algorithm dealing with cases of track drift and re-entry into the FOV individually, in order to match the identities of lost and reappeared targets through a comparison of the affinities between their appearance, size and position, and also to update the status of re-identified targets through this assessment. The experimental results demonstrate that this framework can effectively handle complex temporary lost and re-entry situations with robustness, and that its performance is state of the art.

UR - http://www.scopus.com/inward/record.url?scp=85039911919&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85039911919&partnerID=8YFLogxK

U2 - 10.1109/AVSS.2017.8078505

DO - 10.1109/AVSS.2017.8078505

M3 - Conference contribution

AN - SCOPUS:85039911919

BT - 2017 14th IEEE International Conference on Advanced Video and Signal Based Surveillance, AVSS 2017

PB - Institute of Electrical and Electronics Engineers Inc.

ER -