Unconstrained face verification using fisher vectors computed from frontalized faces

Jun Cheng Chen; Swami Sankaranarayanan; Vishal M. Patel; Rama Chellappa

doi:10.1109/BTAS.2015.7358802

Unconstrained face verification using fisher vectors computed from frontalized faces

Jun Cheng Chen, Swami Sankaranarayanan, Vishal M. Patel, Rama Chellappa

School of Engineering, Electrical & Computer Engineering

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

13 Scopus citations

Abstract

We present an algorithm for unconstrained face verification using Fisher vectors computed from frontalized off-frontal gallery and probe faces. In the training phase, we use the Labeled Faces in the Wild (LFW) dataset to learn the Fisher vector encoding and the joint Bayesian metric. Given an image containing the query face, we perform face detection and landmark localization followed by frontalization to normalize the effect of pose. We further extract dense SIFT features which are then encoded using the Fisher vector learnt during the training phase. The similarity scores are then computed using the learnt joint Bayesian metric. CMC curves and FAR/TAR numbers calculated for a subset of the IARPA JANUS challenge dataset are presented.

Original language	English (US)
Title of host publication	2015 IEEE 7th International Conference on Biometrics Theory, Applications and Systems, BTAS 2015
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9781479987764
DOIs	https://doi.org/10.1109/BTAS.2015.7358802
State	Published - Dec 16 2015
Event	7th IEEE International Conference on Biometrics Theory, Applications and Systems, BTAS 2015 - Arlington, United States Duration: Sep 8 2015 → Sep 11 2015

Publication series

Name	2015 IEEE 7th International Conference on Biometrics Theory, Applications and Systems, BTAS 2015

Other

Other	7th IEEE International Conference on Biometrics Theory, Applications and Systems, BTAS 2015
Country/Territory	United States
City	Arlington
Period	9/8/15 → 9/11/15

All Science Journal Classification (ASJC) codes

Statistics and Probability
Computer Science Applications
Biomedical Engineering

Access to Document

10.1109/BTAS.2015.7358802

Cite this

Chen, J. C., Sankaranarayanan, S., Patel, V. M., & Chellappa, R. (2015). Unconstrained face verification using fisher vectors computed from frontalized faces. In 2015 IEEE 7th International Conference on Biometrics Theory, Applications and Systems, BTAS 2015 Article 7358802 (2015 IEEE 7th International Conference on Biometrics Theory, Applications and Systems, BTAS 2015). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/BTAS.2015.7358802

Chen, Jun Cheng ; Sankaranarayanan, Swami ; Patel, Vishal M. et al. / Unconstrained face verification using fisher vectors computed from frontalized faces. 2015 IEEE 7th International Conference on Biometrics Theory, Applications and Systems, BTAS 2015. Institute of Electrical and Electronics Engineers Inc., 2015. (2015 IEEE 7th International Conference on Biometrics Theory, Applications and Systems, BTAS 2015).

@inproceedings{e4f2eb0915134869abce014765ba0bf1,

title = "Unconstrained face verification using fisher vectors computed from frontalized faces",

abstract = "We present an algorithm for unconstrained face verification using Fisher vectors computed from frontalized off-frontal gallery and probe faces. In the training phase, we use the Labeled Faces in the Wild (LFW) dataset to learn the Fisher vector encoding and the joint Bayesian metric. Given an image containing the query face, we perform face detection and landmark localization followed by frontalization to normalize the effect of pose. We further extract dense SIFT features which are then encoded using the Fisher vector learnt during the training phase. The similarity scores are then computed using the learnt joint Bayesian metric. CMC curves and FAR/TAR numbers calculated for a subset of the IARPA JANUS challenge dataset are presented.",

author = "Chen, {Jun Cheng} and Swami Sankaranarayanan and Patel, {Vishal M.} and Rama Chellappa",

note = "Publisher Copyright: {\textcopyright} 2015 IEEE.; 7th IEEE International Conference on Biometrics Theory, Applications and Systems, BTAS 2015 ; Conference date: 08-09-2015 Through 11-09-2015",

year = "2015",

month = dec,

day = "16",

doi = "10.1109/BTAS.2015.7358802",

language = "English (US)",

series = "2015 IEEE 7th International Conference on Biometrics Theory, Applications and Systems, BTAS 2015",

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

booktitle = "2015 IEEE 7th International Conference on Biometrics Theory, Applications and Systems, BTAS 2015",

address = "United States",

}

Chen, JC, Sankaranarayanan, S, Patel, VM & Chellappa, R 2015, Unconstrained face verification using fisher vectors computed from frontalized faces. in 2015 IEEE 7th International Conference on Biometrics Theory, Applications and Systems, BTAS 2015., 7358802, 2015 IEEE 7th International Conference on Biometrics Theory, Applications and Systems, BTAS 2015, Institute of Electrical and Electronics Engineers Inc., 7th IEEE International Conference on Biometrics Theory, Applications and Systems, BTAS 2015, Arlington, United States, 9/8/15. https://doi.org/10.1109/BTAS.2015.7358802

Unconstrained face verification using fisher vectors computed from frontalized faces. / Chen, Jun Cheng; Sankaranarayanan, Swami; Patel, Vishal M. et al.
2015 IEEE 7th International Conference on Biometrics Theory, Applications and Systems, BTAS 2015. Institute of Electrical and Electronics Engineers Inc., 2015. 7358802 (2015 IEEE 7th International Conference on Biometrics Theory, Applications and Systems, BTAS 2015).

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

TY - GEN

T1 - Unconstrained face verification using fisher vectors computed from frontalized faces

AU - Chen, Jun Cheng

AU - Sankaranarayanan, Swami

AU - Patel, Vishal M.

AU - Chellappa, Rama

PY - 2015/12/16

Y1 - 2015/12/16

N2 - We present an algorithm for unconstrained face verification using Fisher vectors computed from frontalized off-frontal gallery and probe faces. In the training phase, we use the Labeled Faces in the Wild (LFW) dataset to learn the Fisher vector encoding and the joint Bayesian metric. Given an image containing the query face, we perform face detection and landmark localization followed by frontalization to normalize the effect of pose. We further extract dense SIFT features which are then encoded using the Fisher vector learnt during the training phase. The similarity scores are then computed using the learnt joint Bayesian metric. CMC curves and FAR/TAR numbers calculated for a subset of the IARPA JANUS challenge dataset are presented.

AB - We present an algorithm for unconstrained face verification using Fisher vectors computed from frontalized off-frontal gallery and probe faces. In the training phase, we use the Labeled Faces in the Wild (LFW) dataset to learn the Fisher vector encoding and the joint Bayesian metric. Given an image containing the query face, we perform face detection and landmark localization followed by frontalization to normalize the effect of pose. We further extract dense SIFT features which are then encoded using the Fisher vector learnt during the training phase. The similarity scores are then computed using the learnt joint Bayesian metric. CMC curves and FAR/TAR numbers calculated for a subset of the IARPA JANUS challenge dataset are presented.

UR - http://www.scopus.com/inward/record.url?scp=84962855648&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84962855648&partnerID=8YFLogxK

U2 - 10.1109/BTAS.2015.7358802

DO - 10.1109/BTAS.2015.7358802

M3 - Conference contribution

AN - SCOPUS:84962855648

T3 - 2015 IEEE 7th International Conference on Biometrics Theory, Applications and Systems, BTAS 2015

BT - 2015 IEEE 7th International Conference on Biometrics Theory, Applications and Systems, BTAS 2015

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 7th IEEE International Conference on Biometrics Theory, Applications and Systems, BTAS 2015

Y2 - 8 September 2015 through 11 September 2015

ER -

Chen JC, Sankaranarayanan S, Patel VM, Chellappa R. Unconstrained face verification using fisher vectors computed from frontalized faces. In 2015 IEEE 7th International Conference on Biometrics Theory, Applications and Systems, BTAS 2015. Institute of Electrical and Electronics Engineers Inc. 2015. 7358802. (2015 IEEE 7th International Conference on Biometrics Theory, Applications and Systems, BTAS 2015). doi: 10.1109/BTAS.2015.7358802

Unconstrained face verification using fisher vectors computed from frontalized faces

Abstract

Publication series

Other

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this