Data-Augmentation for Reducing Dataset Bias in Person Re-identification

Niall McLaughlin; Jesus Martinez del Rincon; Paul Miller

doi:10.1109/AVSS.2015.7301739

Data-Augmentation for Reducing Dataset Bias in Person Re-identification

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

91 Citations (Scopus)

2046 Downloads (Pure)

Abstract

In this paper we explore ways to address the issue of dataset bias in person re-identification by using data augmentation to increase the variability of the available datasets, and we introduce a novel data augmentation method for re-identification based on changing the image background. We show that use of data augmentation can improve the cross-dataset generalisation of convolutional network based re-identification systems, and that changing the image background yields further improvements.

Original language	English
Title of host publication	Proceedings of 2015 12th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS)
Publisher	Institute of Electrical and Electronics Engineers Inc.
Number of pages	6
ISBN (Print)	9781467376327
DOIs	https://doi.org/10.1109/AVSS.2015.7301739
Publication status	Published - Aug 2015
Event	3rd AMMDS Workshop - AVSS 2015: Activity Monitoring by Multiple Distributed Sensing - Karlsruhe, Germany Duration: 25 Aug 2015 → …

Workshop

Workshop	3rd AMMDS Workshop - AVSS 2015: Activity Monitoring by Multiple Distributed Sensing
Country/Territory	Germany
City	Karlsruhe
Period	25/08/2015 → …

Access to Document

10.1109/AVSS.2015.7301739

Data-Augmentation for Reducing Dataset Bias in Person Re-identification
Copyright 2016 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other users, including reprinting/ republishing this material for advertising or promotional purposes, creating new collective works for resale or redistribution to servers or lists, or reuse of any copyrighted components of this work in other works.
Accepted author manuscript, 3.12 MB

Cite this

@inproceedings{73b03f2351564507b7d51d4c9a5048ef,

title = "Data-Augmentation for Reducing Dataset Bias in Person Re-identification",

abstract = "In this paper we explore ways to address the issue of dataset bias in person re-identification by using data augmentation to increase the variability of the available datasets, and we introduce a novel data augmentation method for re-identification based on changing the image background. We show that use of data augmentation can improve the cross-dataset generalisation of convolutional network based re-identification systems, and that changing the image background yields further improvements.",

author = "Niall McLaughlin and \{Martinez del Rincon\}, Jesus and Paul Miller",

year = "2015",

month = aug,

doi = "10.1109/AVSS.2015.7301739",

language = "English",

isbn = "9781467376327",

booktitle = "Proceedings of 2015 12th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS)",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

address = "United States",

note = " 3rd AMMDS Workshop - AVSS 2015: Activity Monitoring by Multiple Distributed Sensing ; Conference date: 25-08-2015",

}

McLaughlin, N , Martinez del Rincon, J & Miller, P 2015, Data-Augmentation for Reducing Dataset Bias in Person Re-identification. in Proceedings of 2015 12th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS). Institute of Electrical and Electronics Engineers Inc., 3rd AMMDS Workshop - AVSS 2015: Activity Monitoring by Multiple Distributed Sensing, Karlsruhe, Germany, 25/08/2015. https://doi.org/10.1109/AVSS.2015.7301739

Data-Augmentation for Reducing Dataset Bias in Person Re-identification. / McLaughlin, Niall ; Martinez del Rincon, Jesus ; Miller, Paul.
Proceedings of 2015 12th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS). Institute of Electrical and Electronics Engineers Inc., 2015.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Data-Augmentation for Reducing Dataset Bias in Person Re-identification

AU - McLaughlin, Niall

AU - Martinez del Rincon, Jesus

AU - Miller, Paul

PY - 2015/8

Y1 - 2015/8

N2 - In this paper we explore ways to address the issue of dataset bias in person re-identification by using data augmentation to increase the variability of the available datasets, and we introduce a novel data augmentation method for re-identification based on changing the image background. We show that use of data augmentation can improve the cross-dataset generalisation of convolutional network based re-identification systems, and that changing the image background yields further improvements.

AB - In this paper we explore ways to address the issue of dataset bias in person re-identification by using data augmentation to increase the variability of the available datasets, and we introduce a novel data augmentation method for re-identification based on changing the image background. We show that use of data augmentation can improve the cross-dataset generalisation of convolutional network based re-identification systems, and that changing the image background yields further improvements.

UR - http://www.ino.it/ammds/index.html

UR - http://avss2015.org/about-avss/

U2 - 10.1109/AVSS.2015.7301739

DO - 10.1109/AVSS.2015.7301739

M3 - Conference contribution

SN - 9781467376327

BT - Proceedings of 2015 12th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS)

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 3rd AMMDS Workshop - AVSS 2015: Activity Monitoring by Multiple Distributed Sensing

Y2 - 25 August 2015

ER -

Data-Augmentation for Reducing Dataset Bias in Person Re-identification

Abstract

Workshop

Access to Document

Other files and links

Fingerprint

Niall McLaughlin

Cite this

Data-Augmentation for Reducing Dataset Bias in Person Re-identification

Abstract

Workshop

Access to Document

Other files and links

Fingerprint

Profiles

Niall McLaughlin

Cite this