Using nasal curves matching for expression robust 3D nose recognition

M Emambakhsh, A N Evans, Melvyn Smith

Research output: Contribution to conferencePaper

9 Citations (Scopus)

Abstract

The development of 3D face recognition algorithms that are robust to variations in expression has been a challenge for researchers over the past decade. One approach to this problem is to utilize the most stable parts on the face surface. The nasal region's relatively constant structure over various expressions makes it attractive for robust recognition and, in this paper, the use of features from the three-dimensional shape of nose is evaluated. After denoising, face cropping and alignment, the nose region is cropped and 16 landmarks robustly detected on its surface. Pairs of landmarks are connected, which results in 75 curves on the nasal surface; these curves form the feature set. The most stable curves over different expressions and occlusions (due to glasses) are selected and used for nose recognition. The Bosphorus dataset is used for feature selection and FRGC v2.0 for nose recognition. Results show a rank-one recognition rate of 82.58% using only two training samples with varying expression for 505 different subjects and 4879 samples, and 90.87% and 81.61% using the FRGC v2.0 dataset Spring2003 folder for training and the Fall2003 and Spring2004 folders for testing, for neutral and varying expressions, respectively. Given the relatively low complexity of the features and matching, the varying expression results are encouraging.

Conference

Conference6th IEEE International Conference on Biometrics: Theory, Applications and Systems (BTAS 2013)
CountryUSA United States
CityWashington, DC
Period29/09/132/10/13

Fingerprint

Face recognition
Feature extraction
Glass
Testing

Cite this

Emambakhsh, M., Evans, A. N., & Smith , M. (2013). Using nasal curves matching for expression robust 3D nose recognition. 1-8. Paper presented at 6th IEEE International Conference on Biometrics: Theory, Applications and Systems (BTAS 2013), Washington, DC, USA United States. https://doi.org/10.1109/BTAS.2013.6712732

Using nasal curves matching for expression robust 3D nose recognition. / Emambakhsh, M; Evans, A N; Smith , Melvyn.

2013. 1-8 Paper presented at 6th IEEE International Conference on Biometrics: Theory, Applications and Systems (BTAS 2013), Washington, DC, USA United States.

Research output: Contribution to conferencePaper

Emambakhsh, M, Evans, AN & Smith , M 2013, 'Using nasal curves matching for expression robust 3D nose recognition' Paper presented at 6th IEEE International Conference on Biometrics: Theory, Applications and Systems (BTAS 2013), Washington, DC, USA United States, 29/09/13 - 2/10/13, pp. 1-8. https://doi.org/10.1109/BTAS.2013.6712732
Emambakhsh M, Evans AN, Smith M. Using nasal curves matching for expression robust 3D nose recognition. 2013. Paper presented at 6th IEEE International Conference on Biometrics: Theory, Applications and Systems (BTAS 2013), Washington, DC, USA United States. https://doi.org/10.1109/BTAS.2013.6712732
Emambakhsh, M ; Evans, A N ; Smith , Melvyn. / Using nasal curves matching for expression robust 3D nose recognition. Paper presented at 6th IEEE International Conference on Biometrics: Theory, Applications and Systems (BTAS 2013), Washington, DC, USA United States.8 p.
@conference{58102902d477472bbb8941a54e420bd3,
title = "Using nasal curves matching for expression robust 3D nose recognition",
abstract = "The development of 3D face recognition algorithms that are robust to variations in expression has been a challenge for researchers over the past decade. One approach to this problem is to utilize the most stable parts on the face surface. The nasal region's relatively constant structure over various expressions makes it attractive for robust recognition and, in this paper, the use of features from the three-dimensional shape of nose is evaluated. After denoising, face cropping and alignment, the nose region is cropped and 16 landmarks robustly detected on its surface. Pairs of landmarks are connected, which results in 75 curves on the nasal surface; these curves form the feature set. The most stable curves over different expressions and occlusions (due to glasses) are selected and used for nose recognition. The Bosphorus dataset is used for feature selection and FRGC v2.0 for nose recognition. Results show a rank-one recognition rate of 82.58{\%} using only two training samples with varying expression for 505 different subjects and 4879 samples, and 90.87{\%} and 81.61{\%} using the FRGC v2.0 dataset Spring2003 folder for training and the Fall2003 and Spring2004 folders for testing, for neutral and varying expressions, respectively. Given the relatively low complexity of the features and matching, the varying expression results are encouraging.",
author = "M Emambakhsh and Evans, {A N} and Melvyn Smith",
year = "2013",
doi = "10.1109/BTAS.2013.6712732",
language = "English",
pages = "1--8",
note = "6th IEEE International Conference on Biometrics: Theory, Applications and Systems (BTAS 2013) ; Conference date: 29-09-2013 Through 02-10-2013",

}

TY - CONF

T1 - Using nasal curves matching for expression robust 3D nose recognition

AU - Emambakhsh, M

AU - Evans, A N

AU - Smith , Melvyn

PY - 2013

Y1 - 2013

N2 - The development of 3D face recognition algorithms that are robust to variations in expression has been a challenge for researchers over the past decade. One approach to this problem is to utilize the most stable parts on the face surface. The nasal region's relatively constant structure over various expressions makes it attractive for robust recognition and, in this paper, the use of features from the three-dimensional shape of nose is evaluated. After denoising, face cropping and alignment, the nose region is cropped and 16 landmarks robustly detected on its surface. Pairs of landmarks are connected, which results in 75 curves on the nasal surface; these curves form the feature set. The most stable curves over different expressions and occlusions (due to glasses) are selected and used for nose recognition. The Bosphorus dataset is used for feature selection and FRGC v2.0 for nose recognition. Results show a rank-one recognition rate of 82.58% using only two training samples with varying expression for 505 different subjects and 4879 samples, and 90.87% and 81.61% using the FRGC v2.0 dataset Spring2003 folder for training and the Fall2003 and Spring2004 folders for testing, for neutral and varying expressions, respectively. Given the relatively low complexity of the features and matching, the varying expression results are encouraging.

AB - The development of 3D face recognition algorithms that are robust to variations in expression has been a challenge for researchers over the past decade. One approach to this problem is to utilize the most stable parts on the face surface. The nasal region's relatively constant structure over various expressions makes it attractive for robust recognition and, in this paper, the use of features from the three-dimensional shape of nose is evaluated. After denoising, face cropping and alignment, the nose region is cropped and 16 landmarks robustly detected on its surface. Pairs of landmarks are connected, which results in 75 curves on the nasal surface; these curves form the feature set. The most stable curves over different expressions and occlusions (due to glasses) are selected and used for nose recognition. The Bosphorus dataset is used for feature selection and FRGC v2.0 for nose recognition. Results show a rank-one recognition rate of 82.58% using only two training samples with varying expression for 505 different subjects and 4879 samples, and 90.87% and 81.61% using the FRGC v2.0 dataset Spring2003 folder for training and the Fall2003 and Spring2004 folders for testing, for neutral and varying expressions, respectively. Given the relatively low complexity of the features and matching, the varying expression results are encouraging.

UR - http://dx.doi.org/10.1109/BTAS.2013.6712732

UR - http://www.btas2013.org/

U2 - 10.1109/BTAS.2013.6712732

DO - 10.1109/BTAS.2013.6712732

M3 - Paper

SP - 1

EP - 8

ER -