Multimodal Person Re-identification Using RGB-D Sensors and a Transient Identification Database

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

12 Citations (Scopus)
273 Downloads (Pure)

Abstract

This paper describes a system for person re-identification using RGB-D sensors. The system covers the full flow, from detection of subjects, over contour extraction, to re-identification using soft biometrics. The biometrics in question are part-based color histograms and the subjects height. Subjects are added to a transient database and re-identified based on the distance between recorded biometrics and the currently measured metrics. The system works on live video and requires no collaboration from the subjects. The system achieves a 68% re-identification rate with no wrong re-identifications, a result that compares favorable with commercial systems as well as other very recent multimodal re-identification systems.
Original languageEnglish
Title of host publicationInternational Workshop on Biometrics and Forensics
Place of PublicationLisbon
PublisherIEEE Press
Publication date2013
Pages1-4
ISBN (Print)978-1-4673-4987-1
DOIs
Publication statusPublished - 2013
EventInternational Workshop on Biometrics and Forensics - Lisbon, Portugal
Duration: 4 Apr 20135 Apr 2013
Conference number: 1

Workshop

WorkshopInternational Workshop on Biometrics and Forensics
Number1
CountryPortugal
CityLisbon
Period04/04/201305/04/2013

Fingerprint

Biometrics
Sensors
Identification (control systems)
Color

Cite this

@inproceedings{34032309683e49c0b3c9885117e695c3,
title = "Multimodal Person Re-identification Using RGB-D Sensors and a Transient Identification Database",
abstract = "This paper describes a system for person re-identification using RGB-D sensors. The system covers the full flow, from detection of subjects, over contour extraction, to re-identification using soft biometrics. The biometrics in question are part-based color histograms and the subjects height. Subjects are added to a transient database and re-identified based on the distance between recorded biometrics and the currently measured metrics. The system works on live video and requires no collaboration from the subjects. The system achieves a 68{\%} re-identification rate with no wrong re-identifications, a result that compares favorable with commercial systems as well as other very recent multimodal re-identification systems.",
author = "Andreas M{\o}gelmose and Moeslund, {Thomas B.} and Kamal Nasrollahi",
year = "2013",
doi = "10.1109/IWBF.2013.6547322",
language = "English",
isbn = "978-1-4673-4987-1",
pages = "1--4",
booktitle = "International Workshop on Biometrics and Forensics",
publisher = "IEEE Press",

}

Møgelmose, A, Moeslund, TB & Nasrollahi, K 2013, Multimodal Person Re-identification Using RGB-D Sensors and a Transient Identification Database. in International Workshop on Biometrics and Forensics. IEEE Press, Lisbon, pp. 1-4, International Workshop on Biometrics and Forensics, Lisbon, Portugal, 04/04/2013. https://doi.org/10.1109/IWBF.2013.6547322

Multimodal Person Re-identification Using RGB-D Sensors and a Transient Identification Database. / Møgelmose, Andreas; Moeslund, Thomas B.; Nasrollahi, Kamal.

International Workshop on Biometrics and Forensics. Lisbon : IEEE Press, 2013. p. 1-4.

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

TY - GEN

T1 - Multimodal Person Re-identification Using RGB-D Sensors and a Transient Identification Database

AU - Møgelmose, Andreas

AU - Moeslund, Thomas B.

AU - Nasrollahi, Kamal

PY - 2013

Y1 - 2013

N2 - This paper describes a system for person re-identification using RGB-D sensors. The system covers the full flow, from detection of subjects, over contour extraction, to re-identification using soft biometrics. The biometrics in question are part-based color histograms and the subjects height. Subjects are added to a transient database and re-identified based on the distance between recorded biometrics and the currently measured metrics. The system works on live video and requires no collaboration from the subjects. The system achieves a 68% re-identification rate with no wrong re-identifications, a result that compares favorable with commercial systems as well as other very recent multimodal re-identification systems.

AB - This paper describes a system for person re-identification using RGB-D sensors. The system covers the full flow, from detection of subjects, over contour extraction, to re-identification using soft biometrics. The biometrics in question are part-based color histograms and the subjects height. Subjects are added to a transient database and re-identified based on the distance between recorded biometrics and the currently measured metrics. The system works on live video and requires no collaboration from the subjects. The system achieves a 68% re-identification rate with no wrong re-identifications, a result that compares favorable with commercial systems as well as other very recent multimodal re-identification systems.

U2 - 10.1109/IWBF.2013.6547322

DO - 10.1109/IWBF.2013.6547322

M3 - Article in proceeding

SN - 978-1-4673-4987-1

SP - 1

EP - 4

BT - International Workshop on Biometrics and Forensics

PB - IEEE Press

CY - Lisbon

ER -