Utilizing Template Diversity for Fusion of Face Recognizers

Sergey Tulyakov; Nishant Sankaran; Srirangaraj Setlur; Venu Govindaraju

doi:10.1109/ISBA.2019.8778556

Utilizing Template Diversity for Fusion of Face Recognizers

Sergey Tulyakov
, Nishant Sankaran
, Srirangaraj Setlur
, Venu Govindaraju

Research and Economic Development

SUNY Buffalo

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

2 Scopus citations

Abstract

If multiple face images are available for the creation of person's biometric template, some averaging method could be used to combine the feature vectors extracted from each image into a single template feature vector. Resulting average feature vector does not retain the information about image feature vector distribution. In this paper we consider the augmentation of such templates by the information about diversity of constituent face images, e.g. sample standard deviation of image feature vectors. We consider the theoretical model describing the conditions of the usefulness of template diversity measure, and see if such conditions hold in real life templates. We perform our experiments using IARPA face image datasets and deep CNN face recognizers.

Original language	English
Title of host publication	ISBA 2019 - 5th IEEE International Conference on Identity, Security and Behavior Analysis
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9781728105321
DOIs	https://doi.org/10.1109/ISBA.2019.8778556
State	Published - Jan 2019
Event	5th IEEE International Conference on Identity, Security and Behavior Analysis, ISBA 2019 - Hyderabad, India Duration: Jan 22 2019 → Jan 24 2019

Publication series

Name	ISBA 2019 - 5th IEEE International Conference on Identity, Security and Behavior Analysis

Conference

Conference	5th IEEE International Conference on Identity, Security and Behavior Analysis, ISBA 2019
Country/Territory	India
City	Hyderabad
Period	01/22/19 → 01/24/19

Access to Document

10.1109/ISBA.2019.8778556

Cite this

Tulyakov, S., Sankaran, N., Setlur, S., & Govindaraju, V. (2019). Utilizing Template Diversity for Fusion of Face Recognizers. In ISBA 2019 - 5th IEEE International Conference on Identity, Security and Behavior Analysis Article 8778556 (ISBA 2019 - 5th IEEE International Conference on Identity, Security and Behavior Analysis). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ISBA.2019.8778556

Tulyakov, Sergey ; Sankaran, Nishant ; Setlur, Srirangaraj et al. / Utilizing Template Diversity for Fusion of Face Recognizers. ISBA 2019 - 5th IEEE International Conference on Identity, Security and Behavior Analysis. Institute of Electrical and Electronics Engineers Inc., 2019. (ISBA 2019 - 5th IEEE International Conference on Identity, Security and Behavior Analysis).

@inproceedings{61385000fcad402da44c719b4b16a018,

title = "Utilizing Template Diversity for Fusion of Face Recognizers",

abstract = "If multiple face images are available for the creation of person's biometric template, some averaging method could be used to combine the feature vectors extracted from each image into a single template feature vector. Resulting average feature vector does not retain the information about image feature vector distribution. In this paper we consider the augmentation of such templates by the information about diversity of constituent face images, e.g. sample standard deviation of image feature vectors. We consider the theoretical model describing the conditions of the usefulness of template diversity measure, and see if such conditions hold in real life templates. We perform our experiments using IARPA face image datasets and deep CNN face recognizers.",

author = "Sergey Tulyakov and Nishant Sankaran and Srirangaraj Setlur and Venu Govindaraju",

note = "Publisher Copyright: {\textcopyright} 2019 IEEE.; 5th IEEE International Conference on Identity, Security and Behavior Analysis, ISBA 2019 ; Conference date: 22-01-2019 Through 24-01-2019",

year = "2019",

month = jan,

doi = "10.1109/ISBA.2019.8778556",

language = "English",

series = "ISBA 2019 - 5th IEEE International Conference on Identity, Security and Behavior Analysis",

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

booktitle = "ISBA 2019 - 5th IEEE International Conference on Identity, Security and Behavior Analysis",

address = "United States",

}

Tulyakov, S, Sankaran, N, Setlur, S & Govindaraju, V 2019, Utilizing Template Diversity for Fusion of Face Recognizers. in ISBA 2019 - 5th IEEE International Conference on Identity, Security and Behavior Analysis., 8778556, ISBA 2019 - 5th IEEE International Conference on Identity, Security and Behavior Analysis, Institute of Electrical and Electronics Engineers Inc., 5th IEEE International Conference on Identity, Security and Behavior Analysis, ISBA 2019, Hyderabad, India, 01/22/19. https://doi.org/10.1109/ISBA.2019.8778556

Utilizing Template Diversity for Fusion of Face Recognizers. / Tulyakov, Sergey; Sankaran, Nishant; Setlur, Srirangaraj et al.
ISBA 2019 - 5th IEEE International Conference on Identity, Security and Behavior Analysis. Institute of Electrical and Electronics Engineers Inc., 2019. 8778556 (ISBA 2019 - 5th IEEE International Conference on Identity, Security and Behavior Analysis).

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

TY - GEN

T1 - Utilizing Template Diversity for Fusion of Face Recognizers

AU - Tulyakov, Sergey

AU - Sankaran, Nishant

AU - Setlur, Srirangaraj

AU - Govindaraju, Venu

PY - 2019/1

Y1 - 2019/1

N2 - If multiple face images are available for the creation of person's biometric template, some averaging method could be used to combine the feature vectors extracted from each image into a single template feature vector. Resulting average feature vector does not retain the information about image feature vector distribution. In this paper we consider the augmentation of such templates by the information about diversity of constituent face images, e.g. sample standard deviation of image feature vectors. We consider the theoretical model describing the conditions of the usefulness of template diversity measure, and see if such conditions hold in real life templates. We perform our experiments using IARPA face image datasets and deep CNN face recognizers.

AB - If multiple face images are available for the creation of person's biometric template, some averaging method could be used to combine the feature vectors extracted from each image into a single template feature vector. Resulting average feature vector does not retain the information about image feature vector distribution. In this paper we consider the augmentation of such templates by the information about diversity of constituent face images, e.g. sample standard deviation of image feature vectors. We consider the theoretical model describing the conditions of the usefulness of template diversity measure, and see if such conditions hold in real life templates. We perform our experiments using IARPA face image datasets and deep CNN face recognizers.

UR - https://www.scopus.com/pages/publications/85070643421

U2 - 10.1109/ISBA.2019.8778556

DO - 10.1109/ISBA.2019.8778556

M3 - Conference contribution

AN - SCOPUS:85070643421

T3 - ISBA 2019 - 5th IEEE International Conference on Identity, Security and Behavior Analysis

BT - ISBA 2019 - 5th IEEE International Conference on Identity, Security and Behavior Analysis

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 5th IEEE International Conference on Identity, Security and Behavior Analysis, ISBA 2019

Y2 - 22 January 2019 through 24 January 2019

ER -

Tulyakov S, Sankaran N, Setlur S , Govindaraju V. Utilizing Template Diversity for Fusion of Face Recognizers. In ISBA 2019 - 5th IEEE International Conference on Identity, Security and Behavior Analysis. Institute of Electrical and Electronics Engineers Inc. 2019. 8778556. (ISBA 2019 - 5th IEEE International Conference on Identity, Security and Behavior Analysis). doi: 10.1109/ISBA.2019.8778556

Utilizing Template Diversity for Fusion of Face Recognizers

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this