Home > Research > Publications & Outputs > Discriminant Analysis via Joint Euler Transform...

Electronic data

  • Euler-LDA-L21_TIP

    Rights statement: ©2018 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE.

    Accepted author manuscript, 750 KB, PDF document

    Available under license: CC BY-NC: Creative Commons Attribution-NonCommercial 4.0 International License

Links

Text available via DOI:

View graph of relations

Discriminant Analysis via Joint Euler Transform and ℓ2, 1-Norm

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Published

Standard

Discriminant Analysis via Joint Euler Transform and ℓ2, 1-Norm. / Liao, Shuangli; Gao, Quanxue; Yang, Zhaohua et al.
In: IEEE Transactions on Image Processing, Vol. 27, No. 11, 11.2018, p. 5668-5682.

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Harvard

Liao, S, Gao, Q, Yang, Z, Chen, F, Nie, F & Han, J 2018, 'Discriminant Analysis via Joint Euler Transform and ℓ2, 1-Norm', IEEE Transactions on Image Processing, vol. 27, no. 11, pp. 5668-5682. https://doi.org/10.1109/TIP.2018.2859589

APA

Liao, S., Gao, Q., Yang, Z., Chen, F., Nie, F., & Han, J. (2018). Discriminant Analysis via Joint Euler Transform and ℓ2, 1-Norm. IEEE Transactions on Image Processing, 27(11), 5668-5682. https://doi.org/10.1109/TIP.2018.2859589

Vancouver

Liao S, Gao Q, Yang Z, Chen F, Nie F, Han J. Discriminant Analysis via Joint Euler Transform and ℓ2, 1-Norm. IEEE Transactions on Image Processing. 2018 Nov;27(11):5668-5682. Epub 2018 Jul 25. doi: 10.1109/TIP.2018.2859589

Author

Liao, Shuangli ; Gao, Quanxue ; Yang, Zhaohua et al. / Discriminant Analysis via Joint Euler Transform and ℓ2, 1-Norm. In: IEEE Transactions on Image Processing. 2018 ; Vol. 27, No. 11. pp. 5668-5682.

Bibtex

@article{078cc5823e844aabac81d160120dac51,
title = "Discriminant Analysis via Joint Euler Transform and ℓ2, 1-Norm",
abstract = "Linear discriminant analysis (LDA) has been widely used for face recognition. However, when identifying faces in the wild, the existence of outliers that deviate significantly from the rest of the data can arbitrarily skew the desired solution. This usually deteriorates LDA{\textquoteright}s performance dramatically, thus preventing it from mass deployment in real-world applications. To handle this problem, we propose an effective distance metric learning method-based LDA, namely, Euler LDA-L21 (e-LDA-L21). e-LDA-L21 is carried out in two stages, in which each image is mapped into a complex space by Euler transform in the first stage and the ℓ2,1 -norm is adopted as the distance metric in the second stage. This not only reveals nonlinear features but also exploits the geometric structure of data. To solve e-LDA-L21 efficiently, we propose an iterative algorithm, which is a closed-form solution at each iteration with convergence guaranteed. Finally, we extend e-LDA-L21 to Euler 2DLDA-L21 (e-2DLDA-L21) which further exploits the spatial information embedded in image pixels. Experimental results on several face databases demonstrate its superiority over the state-of-the-art algorithms.",
author = "Shuangli Liao and Quanxue Gao and Zhaohua Yang and Fang Chen and Feiping Nie and Jungong Han",
note = "{\textcopyright}2018 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE.",
year = "2018",
month = nov,
doi = "10.1109/TIP.2018.2859589",
language = "English",
volume = "27",
pages = "5668--5682",
journal = "IEEE Transactions on Image Processing",
issn = "1057-7149",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "11",

}

RIS

TY - JOUR

T1 - Discriminant Analysis via Joint Euler Transform and ℓ2, 1-Norm

AU - Liao, Shuangli

AU - Gao, Quanxue

AU - Yang, Zhaohua

AU - Chen, Fang

AU - Nie, Feiping

AU - Han, Jungong

N1 - ©2018 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE.

PY - 2018/11

Y1 - 2018/11

N2 - Linear discriminant analysis (LDA) has been widely used for face recognition. However, when identifying faces in the wild, the existence of outliers that deviate significantly from the rest of the data can arbitrarily skew the desired solution. This usually deteriorates LDA’s performance dramatically, thus preventing it from mass deployment in real-world applications. To handle this problem, we propose an effective distance metric learning method-based LDA, namely, Euler LDA-L21 (e-LDA-L21). e-LDA-L21 is carried out in two stages, in which each image is mapped into a complex space by Euler transform in the first stage and the ℓ2,1 -norm is adopted as the distance metric in the second stage. This not only reveals nonlinear features but also exploits the geometric structure of data. To solve e-LDA-L21 efficiently, we propose an iterative algorithm, which is a closed-form solution at each iteration with convergence guaranteed. Finally, we extend e-LDA-L21 to Euler 2DLDA-L21 (e-2DLDA-L21) which further exploits the spatial information embedded in image pixels. Experimental results on several face databases demonstrate its superiority over the state-of-the-art algorithms.

AB - Linear discriminant analysis (LDA) has been widely used for face recognition. However, when identifying faces in the wild, the existence of outliers that deviate significantly from the rest of the data can arbitrarily skew the desired solution. This usually deteriorates LDA’s performance dramatically, thus preventing it from mass deployment in real-world applications. To handle this problem, we propose an effective distance metric learning method-based LDA, namely, Euler LDA-L21 (e-LDA-L21). e-LDA-L21 is carried out in two stages, in which each image is mapped into a complex space by Euler transform in the first stage and the ℓ2,1 -norm is adopted as the distance metric in the second stage. This not only reveals nonlinear features but also exploits the geometric structure of data. To solve e-LDA-L21 efficiently, we propose an iterative algorithm, which is a closed-form solution at each iteration with convergence guaranteed. Finally, we extend e-LDA-L21 to Euler 2DLDA-L21 (e-2DLDA-L21) which further exploits the spatial information embedded in image pixels. Experimental results on several face databases demonstrate its superiority over the state-of-the-art algorithms.

U2 - 10.1109/TIP.2018.2859589

DO - 10.1109/TIP.2018.2859589

M3 - Journal article

VL - 27

SP - 5668

EP - 5682

JO - IEEE Transactions on Image Processing

JF - IEEE Transactions on Image Processing

SN - 1057-7149

IS - 11

ER -