Robust local features for remote face recognition

Jie Chen, Li Liu, Vili Kellokumpu, Guoying Zhao, Matti Pietikäinen, Vishal M. Patel, Rama Chellappa

Research output: Contribution to journalArticleScientificpeer-review

10 Citations (Scopus)

Abstract

In this paper, we propose a robust local descriptor for face recognition. It consists of two components, one based on a shearlet-decomposition and the other on local binary pattern (LBP). Shearlets can completely analyze the singular structures of piecewise smooth images, which is useful since singularities and irregular structures carry useful information in an underlying image. Furthermore, LBP is effective for describing the edges extracted by shearlets even when the images contain high level of noise. Experimental results using the Face Recognition Grand Challenge dataset show that the proposed local descriptor significantly outperforms many widely used features (e.g., Gabor and deep learning-based features) when the images are corrupted by random noise, demonstrating robustness to noise. In addition, experimental results show promising results for two challenging datasets which have poor image quality, i.e., a remote face dataset and the Point and Shoot Face Recognition Challenge dataset.

Original languageEnglish
Pages (from-to)34-46
Number of pages13
JournalImage and Vision Computing
Volume64
DOIs
Publication statusPublished - Aug 2017
MoE publication typeNot Eligible

Fingerprint

Face recognition
Image quality
Decomposition

Keywords

  • Remote face recognition
  • Robust local feature

Cite this

Chen, J., Liu, L., Kellokumpu, V., Zhao, G., Pietikäinen, M., Patel, V. M., & Chellappa, R. (2017). Robust local features for remote face recognition. Image and Vision Computing, 64, 34-46. https://doi.org/10.1016/j.imavis.2017.05.006
Chen, Jie ; Liu, Li ; Kellokumpu, Vili ; Zhao, Guoying ; Pietikäinen, Matti ; Patel, Vishal M. ; Chellappa, Rama. / Robust local features for remote face recognition. In: Image and Vision Computing. 2017 ; Vol. 64. pp. 34-46.
@article{56cc393084124ef0a2da07e392c3c863,
title = "Robust local features for remote face recognition",
abstract = "In this paper, we propose a robust local descriptor for face recognition. It consists of two components, one based on a shearlet-decomposition and the other on local binary pattern (LBP). Shearlets can completely analyze the singular structures of piecewise smooth images, which is useful since singularities and irregular structures carry useful information in an underlying image. Furthermore, LBP is effective for describing the edges extracted by shearlets even when the images contain high level of noise. Experimental results using the Face Recognition Grand Challenge dataset show that the proposed local descriptor significantly outperforms many widely used features (e.g., Gabor and deep learning-based features) when the images are corrupted by random noise, demonstrating robustness to noise. In addition, experimental results show promising results for two challenging datasets which have poor image quality, i.e., a remote face dataset and the Point and Shoot Face Recognition Challenge dataset.",
keywords = "Remote face recognition, Robust local feature",
author = "Jie Chen and Li Liu and Vili Kellokumpu and Guoying Zhao and Matti Pietik{\"a}inen and Patel, {Vishal M.} and Rama Chellappa",
year = "2017",
month = "8",
doi = "10.1016/j.imavis.2017.05.006",
language = "English",
volume = "64",
pages = "34--46",
journal = "Image and Vision Computing",
issn = "0262-8856",
publisher = "Elsevier",

}

Chen, J, Liu, L, Kellokumpu, V, Zhao, G, Pietikäinen, M, Patel, VM & Chellappa, R 2017, 'Robust local features for remote face recognition', Image and Vision Computing, vol. 64, pp. 34-46. https://doi.org/10.1016/j.imavis.2017.05.006

Robust local features for remote face recognition. / Chen, Jie; Liu, Li; Kellokumpu, Vili; Zhao, Guoying; Pietikäinen, Matti; Patel, Vishal M.; Chellappa, Rama.

In: Image and Vision Computing, Vol. 64, 08.2017, p. 34-46.

Research output: Contribution to journalArticleScientificpeer-review

TY - JOUR

T1 - Robust local features for remote face recognition

AU - Chen, Jie

AU - Liu, Li

AU - Kellokumpu, Vili

AU - Zhao, Guoying

AU - Pietikäinen, Matti

AU - Patel, Vishal M.

AU - Chellappa, Rama

PY - 2017/8

Y1 - 2017/8

N2 - In this paper, we propose a robust local descriptor for face recognition. It consists of two components, one based on a shearlet-decomposition and the other on local binary pattern (LBP). Shearlets can completely analyze the singular structures of piecewise smooth images, which is useful since singularities and irregular structures carry useful information in an underlying image. Furthermore, LBP is effective for describing the edges extracted by shearlets even when the images contain high level of noise. Experimental results using the Face Recognition Grand Challenge dataset show that the proposed local descriptor significantly outperforms many widely used features (e.g., Gabor and deep learning-based features) when the images are corrupted by random noise, demonstrating robustness to noise. In addition, experimental results show promising results for two challenging datasets which have poor image quality, i.e., a remote face dataset and the Point and Shoot Face Recognition Challenge dataset.

AB - In this paper, we propose a robust local descriptor for face recognition. It consists of two components, one based on a shearlet-decomposition and the other on local binary pattern (LBP). Shearlets can completely analyze the singular structures of piecewise smooth images, which is useful since singularities and irregular structures carry useful information in an underlying image. Furthermore, LBP is effective for describing the edges extracted by shearlets even when the images contain high level of noise. Experimental results using the Face Recognition Grand Challenge dataset show that the proposed local descriptor significantly outperforms many widely used features (e.g., Gabor and deep learning-based features) when the images are corrupted by random noise, demonstrating robustness to noise. In addition, experimental results show promising results for two challenging datasets which have poor image quality, i.e., a remote face dataset and the Point and Shoot Face Recognition Challenge dataset.

KW - Remote face recognition

KW - Robust local feature

UR - http://www.scopus.com/inward/record.url?scp=85020919834&partnerID=8YFLogxK

U2 - 10.1016/j.imavis.2017.05.006

DO - 10.1016/j.imavis.2017.05.006

M3 - Article

AN - SCOPUS:85020919834

VL - 64

SP - 34

EP - 46

JO - Image and Vision Computing

JF - Image and Vision Computing

SN - 0262-8856

ER -