Shallow CNNs for the Reliable Detection of Facial Marks

Chris Zeinstra; Erwin Haasnoot

doi:10.23919/BIOSIG.2018.8553157

Shallow CNNs for the Reliable Detection of Facial Marks

Chris Zeinstra, Erwin Haasnoot

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

1 Downloads (Pure)

Abstract

Facial marks are local irregularities of skin texture. Their type and/or spatial pattern can be used as a (soft) biometric modality in several applications. A key requirement for a biometric system that utilises facial marks is their reliable detection. Detection methods typically use a blob detector followed by heuristic post processing steps to reduce the number of false positives. In this paper, we consider shallow Convolutional Neural Networks (CNNs) for facial mark detection. The choice of this network type seems natural as it learns multiple (non) blob detectors; shallow refers to the fact that we only consider CNNs up to three layers. We show that (a) these CNNs successfully address the false positive problem, (b) remove the need for post processing steps, and (c) outperform a classic blob detector, approaches taken in previous studies and some other non CNN type classifiers in terms of EER and FMR at TMR=0.95.

Original language	English
Title of host publication	2018 International Conference of the Biometrics Special Interest Group, BIOSIG 2018
Editors	Arslan Bromme, Andreas Uhl, Christoph Busch, Christian Rathgeb, Antitza Dantcheva
Place of Publication	Piscataway, NJ
Publisher	IEEE
ISBN (Electronic)	978-3-88579-676-3
ISBN (Print)	978-1-5386-6007-2
DOIs	https://doi.org/10.23919/BIOSIG.2018.8553157
Publication status	Published - 10 Oct 2018
Event	17th International Conference of the Biometrics Special Interest Group, BIOSIG 2018 - Darmstadt, Germany Duration: 26 Sep 2018 → 28 Sep 2018 Conference number: 17

Publication series

Name	International Conference of the Biometrics Special Interest Group (BIOSIG)
Publisher	IEEE
Volume	2018
ISSN (Electronic)	1617-5468

Conference

Conference	17th International Conference of the Biometrics Special Interest Group, BIOSIG 2018
Abbreviated title	BIOSIG 2018
Country/Territory	Germany
City	Darmstadt
Period	26/09/18 → 28/09/18

Keywords

CNN
Facial marks
Forensics
Image processing

Access to Document

10.23919/BIOSIG.2018.8553157

Cite this

Zeinstra, C., & Haasnoot, E. (2018). Shallow CNNs for the Reliable Detection of Facial Marks. In A. Bromme, A. Uhl, C. Busch, C. Rathgeb, & A. Dantcheva (Eds.), 2018 International Conference of the Biometrics Special Interest Group, BIOSIG 2018 [8553157] (International Conference of the Biometrics Special Interest Group (BIOSIG); Vol. 2018). IEEE. https://doi.org/10.23919/BIOSIG.2018.8553157

Zeinstra, Chris ; Haasnoot, Erwin. / Shallow CNNs for the Reliable Detection of Facial Marks. 2018 International Conference of the Biometrics Special Interest Group, BIOSIG 2018. editor / Arslan Bromme ; Andreas Uhl ; Christoph Busch ; Christian Rathgeb ; Antitza Dantcheva. Piscataway, NJ : IEEE, 2018. (International Conference of the Biometrics Special Interest Group (BIOSIG)).

@inproceedings{c74e54d346394531a75d76410356aff7,

title = "Shallow CNNs for the Reliable Detection of Facial Marks",

abstract = "Facial marks are local irregularities of skin texture. Their type and/or spatial pattern can be used as a (soft) biometric modality in several applications. A key requirement for a biometric system that utilises facial marks is their reliable detection. Detection methods typically use a blob detector followed by heuristic post processing steps to reduce the number of false positives. In this paper, we consider shallow Convolutional Neural Networks (CNNs) for facial mark detection. The choice of this network type seems natural as it learns multiple (non) blob detectors; shallow refers to the fact that we only consider CNNs up to three layers. We show that (a) these CNNs successfully address the false positive problem, (b) remove the need for post processing steps, and (c) outperform a classic blob detector, approaches taken in previous studies and some other non CNN type classifiers in terms of EER and FMR at TMR=0.95.",

keywords = "CNN, Facial marks, Forensics, Image processing",

author = "Chris Zeinstra and Erwin Haasnoot",

year = "2018",

month = oct,

day = "10",

doi = "10.23919/BIOSIG.2018.8553157",

language = "English",

isbn = "978-1-5386-6007-2",

series = "International Conference of the Biometrics Special Interest Group (BIOSIG)",

publisher = "IEEE",

editor = "Arslan Bromme and Andreas Uhl and Christoph Busch and Christian Rathgeb and Antitza Dantcheva",

booktitle = "2018 International Conference of the Biometrics Special Interest Group, BIOSIG 2018",

address = "United States",

note = "17th International Conference of the Biometrics Special Interest Group, BIOSIG 2018, BIOSIG 2018 ; Conference date: 26-09-2018 Through 28-09-2018",

}

Zeinstra, C & Haasnoot, E 2018, Shallow CNNs for the Reliable Detection of Facial Marks. in A Bromme, A Uhl, C Busch, C Rathgeb & A Dantcheva (eds), 2018 International Conference of the Biometrics Special Interest Group, BIOSIG 2018., 8553157, International Conference of the Biometrics Special Interest Group (BIOSIG), vol. 2018, IEEE, Piscataway, NJ, 17th International Conference of the Biometrics Special Interest Group, BIOSIG 2018, Darmstadt, Hesse, Germany, 26/09/18. https://doi.org/10.23919/BIOSIG.2018.8553157

Shallow CNNs for the Reliable Detection of Facial Marks. / Zeinstra, Chris ; Haasnoot, Erwin.

2018 International Conference of the Biometrics Special Interest Group, BIOSIG 2018. ed. / Arslan Bromme; Andreas Uhl; Christoph Busch; Christian Rathgeb; Antitza Dantcheva. Piscataway, NJ : IEEE, 2018. 8553157 (International Conference of the Biometrics Special Interest Group (BIOSIG); Vol. 2018).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - Shallow CNNs for the Reliable Detection of Facial Marks

AU - Zeinstra, Chris

AU - Haasnoot, Erwin

N1 - Conference code: 17

PY - 2018/10/10

Y1 - 2018/10/10

N2 - Facial marks are local irregularities of skin texture. Their type and/or spatial pattern can be used as a (soft) biometric modality in several applications. A key requirement for a biometric system that utilises facial marks is their reliable detection. Detection methods typically use a blob detector followed by heuristic post processing steps to reduce the number of false positives. In this paper, we consider shallow Convolutional Neural Networks (CNNs) for facial mark detection. The choice of this network type seems natural as it learns multiple (non) blob detectors; shallow refers to the fact that we only consider CNNs up to three layers. We show that (a) these CNNs successfully address the false positive problem, (b) remove the need for post processing steps, and (c) outperform a classic blob detector, approaches taken in previous studies and some other non CNN type classifiers in terms of EER and FMR at TMR=0.95.

AB - Facial marks are local irregularities of skin texture. Their type and/or spatial pattern can be used as a (soft) biometric modality in several applications. A key requirement for a biometric system that utilises facial marks is their reliable detection. Detection methods typically use a blob detector followed by heuristic post processing steps to reduce the number of false positives. In this paper, we consider shallow Convolutional Neural Networks (CNNs) for facial mark detection. The choice of this network type seems natural as it learns multiple (non) blob detectors; shallow refers to the fact that we only consider CNNs up to three layers. We show that (a) these CNNs successfully address the false positive problem, (b) remove the need for post processing steps, and (c) outperform a classic blob detector, approaches taken in previous studies and some other non CNN type classifiers in terms of EER and FMR at TMR=0.95.

KW - CNN

KW - Facial marks

KW - Forensics

KW - Image processing

UR - http://www.scopus.com/inward/record.url?scp=85060026186&partnerID=8YFLogxK

U2 - 10.23919/BIOSIG.2018.8553157

DO - 10.23919/BIOSIG.2018.8553157

M3 - Conference contribution

AN - SCOPUS:85060026186

SN - 978-1-5386-6007-2

T3 - International Conference of the Biometrics Special Interest Group (BIOSIG)

BT - 2018 International Conference of the Biometrics Special Interest Group, BIOSIG 2018

A2 - Bromme, Arslan

A2 - Uhl, Andreas

A2 - Busch, Christoph

A2 - Rathgeb, Christian

A2 - Dantcheva, Antitza

PB - IEEE

CY - Piscataway, NJ

T2 - 17th International Conference of the Biometrics Special Interest Group, BIOSIG 2018

Y2 - 26 September 2018 through 28 September 2018

ER -

Zeinstra C , Haasnoot E. Shallow CNNs for the Reliable Detection of Facial Marks. In Bromme A, Uhl A, Busch C, Rathgeb C, Dantcheva A, editors, 2018 International Conference of the Biometrics Special Interest Group, BIOSIG 2018. Piscataway, NJ: IEEE. 2018. 8553157. (International Conference of the Biometrics Special Interest Group (BIOSIG)). doi: 10.23919/BIOSIG.2018.8553157

Shallow CNNs for the Reliable Detection of Facial Marks

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this