Towards Robust Evaluation of Face Morphing Detection

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

1 Citation (Scopus)
1 Downloads (Pure)

Abstract

Automated face recognition is increasingly used as a reliable means to establish the identity of persons for various purposes, ranging from automated passport checks at the border to transferring money and unlocking mobile phones. Face morphing is a technique to blend facial images of two or more subjects such that the result resembles both subjects. Face morphing attacks pose a serious risk for any face recognition system. Without automated morphing detection, state of the art face recognition systems are extremely vulnerable to morphing attacks. Morphing detection methods published in literature often only work for a few types of morphs or on a single dataset with morphed photographs. We create face morphing databases with varying characteristics and how for a LBP/SVM based morphing detection method that performs on par with the state of the art (around 2% EER), the performance collapses with an EER as high as if it is tested across databases with different characteristics. In addition we show that simple image manipulations like adding noise or rescaling can be used to obscure morphing artifacts and deteriorate the morphing detection performance.
Original languageEnglish
Title of host publication2018 26th European Signal Processing Conference, EUSIPCO 2018
Place of PublicationPiscataway, NJ
PublisherIEEE
Pages1027-1031
Number of pages5
ISBN (Electronic)978-9-0827-9701-5 , 978-90-827970-0-8
ISBN (Print)978-1-5386-3736-4
DOIs
Publication statusPublished - 29 Nov 2018
Event26th European Signal Processing Conference, EUSIPCO 2018 - Rome, Italy
Duration: 3 Sep 20187 Sep 2018
Conference number: 26
http://www.eusipco2018.org/

Publication series

NameEuropean Signal Processing Conference
Volume2018-September
ISSN (Print)2219-5491

Conference

Conference26th European Signal Processing Conference, EUSIPCO 2018
Abbreviated titleEUSIPCO
CountryItaly
CityRome
Period3/09/187/09/18
Internet address

Fingerprint

Face recognition
Mobile phones

Cite this

Spreeuwers, L., Veldhuis, R., & Schils, M. (2018). Towards Robust Evaluation of Face Morphing Detection. In 2018 26th European Signal Processing Conference, EUSIPCO 2018 (pp. 1027-1031). [8553018] (European Signal Processing Conference; Vol. 2018-September). Piscataway, NJ: IEEE. https://doi.org/10.23919/EUSIPCO.2018.8553018
Spreeuwers, Luuk ; Veldhuis, Raymond ; Schils, Maikel. / Towards Robust Evaluation of Face Morphing Detection. 2018 26th European Signal Processing Conference, EUSIPCO 2018. Piscataway, NJ : IEEE, 2018. pp. 1027-1031 (European Signal Processing Conference).
@inproceedings{f3dff3ffc77d4f379efafefac40de5dd,
title = "Towards Robust Evaluation of Face Morphing Detection",
abstract = "Automated face recognition is increasingly used as a reliable means to establish the identity of persons for various purposes, ranging from automated passport checks at the border to transferring money and unlocking mobile phones. Face morphing is a technique to blend facial images of two or more subjects such that the result resembles both subjects. Face morphing attacks pose a serious risk for any face recognition system. Without automated morphing detection, state of the art face recognition systems are extremely vulnerable to morphing attacks. Morphing detection methods published in literature often only work for a few types of morphs or on a single dataset with morphed photographs. We create face morphing databases with varying characteristics and how for a LBP/SVM based morphing detection method that performs on par with the state of the art (around 2{\%} EER), the performance collapses with an EER as high as if it is tested across databases with different characteristics. In addition we show that simple image manipulations like adding noise or rescaling can be used to obscure morphing artifacts and deteriorate the morphing detection performance.",
author = "Luuk Spreeuwers and Raymond Veldhuis and Maikel Schils",
year = "2018",
month = "11",
day = "29",
doi = "10.23919/EUSIPCO.2018.8553018",
language = "English",
isbn = "978-1-5386-3736-4",
series = "European Signal Processing Conference",
publisher = "IEEE",
pages = "1027--1031",
booktitle = "2018 26th European Signal Processing Conference, EUSIPCO 2018",
address = "United States",

}

Spreeuwers, L, Veldhuis, R & Schils, M 2018, Towards Robust Evaluation of Face Morphing Detection. in 2018 26th European Signal Processing Conference, EUSIPCO 2018., 8553018, European Signal Processing Conference, vol. 2018-September, IEEE, Piscataway, NJ, pp. 1027-1031, 26th European Signal Processing Conference, EUSIPCO 2018, Rome, Italy, 3/09/18. https://doi.org/10.23919/EUSIPCO.2018.8553018

Towards Robust Evaluation of Face Morphing Detection. / Spreeuwers, Luuk; Veldhuis, Raymond; Schils, Maikel.

2018 26th European Signal Processing Conference, EUSIPCO 2018. Piscataway, NJ : IEEE, 2018. p. 1027-1031 8553018 (European Signal Processing Conference; Vol. 2018-September).

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

TY - GEN

T1 - Towards Robust Evaluation of Face Morphing Detection

AU - Spreeuwers, Luuk

AU - Veldhuis, Raymond

AU - Schils, Maikel

PY - 2018/11/29

Y1 - 2018/11/29

N2 - Automated face recognition is increasingly used as a reliable means to establish the identity of persons for various purposes, ranging from automated passport checks at the border to transferring money and unlocking mobile phones. Face morphing is a technique to blend facial images of two or more subjects such that the result resembles both subjects. Face morphing attacks pose a serious risk for any face recognition system. Without automated morphing detection, state of the art face recognition systems are extremely vulnerable to morphing attacks. Morphing detection methods published in literature often only work for a few types of morphs or on a single dataset with morphed photographs. We create face morphing databases with varying characteristics and how for a LBP/SVM based morphing detection method that performs on par with the state of the art (around 2% EER), the performance collapses with an EER as high as if it is tested across databases with different characteristics. In addition we show that simple image manipulations like adding noise or rescaling can be used to obscure morphing artifacts and deteriorate the morphing detection performance.

AB - Automated face recognition is increasingly used as a reliable means to establish the identity of persons for various purposes, ranging from automated passport checks at the border to transferring money and unlocking mobile phones. Face morphing is a technique to blend facial images of two or more subjects such that the result resembles both subjects. Face morphing attacks pose a serious risk for any face recognition system. Without automated morphing detection, state of the art face recognition systems are extremely vulnerable to morphing attacks. Morphing detection methods published in literature often only work for a few types of morphs or on a single dataset with morphed photographs. We create face morphing databases with varying characteristics and how for a LBP/SVM based morphing detection method that performs on par with the state of the art (around 2% EER), the performance collapses with an EER as high as if it is tested across databases with different characteristics. In addition we show that simple image manipulations like adding noise or rescaling can be used to obscure morphing artifacts and deteriorate the morphing detection performance.

UR - http://www.scopus.com/inward/record.url?scp=85059807161&partnerID=8YFLogxK

U2 - 10.23919/EUSIPCO.2018.8553018

DO - 10.23919/EUSIPCO.2018.8553018

M3 - Conference contribution

SN - 978-1-5386-3736-4

T3 - European Signal Processing Conference

SP - 1027

EP - 1031

BT - 2018 26th European Signal Processing Conference, EUSIPCO 2018

PB - IEEE

CY - Piscataway, NJ

ER -

Spreeuwers L, Veldhuis R, Schils M. Towards Robust Evaluation of Face Morphing Detection. In 2018 26th European Signal Processing Conference, EUSIPCO 2018. Piscataway, NJ: IEEE. 2018. p. 1027-1031. 8553018. (European Signal Processing Conference). https://doi.org/10.23919/EUSIPCO.2018.8553018