Tampere University of Technology

TUTCRIS Research Portal

Scaling up Class-Specific Kernel Discriminant Analysis for large-scale Face Verification

Research output: Contribution to journalArticleScientificpeer-review

Standard

Scaling up Class-Specific Kernel Discriminant Analysis for large-scale Face Verification. / Iosifidis, Alexandros; Gabbouj, Moncef.

In: IEEE Transactions on Information Forensics and Security, Vol. 11, No. 11, 2016, p. 2453-2465.

Research output: Contribution to journalArticleScientificpeer-review

Harvard

Iosifidis, A & Gabbouj, M 2016, 'Scaling up Class-Specific Kernel Discriminant Analysis for large-scale Face Verification', IEEE Transactions on Information Forensics and Security, vol. 11, no. 11, pp. 2453-2465. https://doi.org/10.1109/TIFS.2016.2582562

APA

Iosifidis, A., & Gabbouj, M. (2016). Scaling up Class-Specific Kernel Discriminant Analysis for large-scale Face Verification. IEEE Transactions on Information Forensics and Security, 11(11), 2453-2465. https://doi.org/10.1109/TIFS.2016.2582562

Vancouver

Iosifidis A, Gabbouj M. Scaling up Class-Specific Kernel Discriminant Analysis for large-scale Face Verification. IEEE Transactions on Information Forensics and Security. 2016;11(11):2453-2465. https://doi.org/10.1109/TIFS.2016.2582562

Author

Iosifidis, Alexandros ; Gabbouj, Moncef. / Scaling up Class-Specific Kernel Discriminant Analysis for large-scale Face Verification. In: IEEE Transactions on Information Forensics and Security. 2016 ; Vol. 11, No. 11. pp. 2453-2465.

Bibtex - Download

@article{8153fa048b7f418fad0d1177c1ae43d4,
title = "Scaling up Class-Specific Kernel Discriminant Analysis for large-scale Face Verification",
abstract = "In this paper, a novel approximate solution of the criterion used in non-linear class-specific discriminant subspace learning is proposed. We build on the Class-Specific Kernel Spectral Regression method which is a two-step process formed by an eigenanalysis step and a kernel regression step. Based on the structure of the intra-class and out-of-class scatter matrices, we provide a fast solution for the first step. For the second step, we propose the use of approximate kernel space definitions. We analytically show that the adoption of randomized and classspecific kernels have the effect of regularization and Nystr¨ombased approximation, respectively. We evaluate the proposed approach in face verification problems and compare it with existing approaches. Experimental results show the effectiveness and efficiency of the proposed Approximate Class-Specific Kernel Spectral Regression method, since it can provide satisfactory performance and scale well with the size of the data.",
author = "Alexandros Iosifidis and Moncef Gabbouj",
year = "2016",
doi = "10.1109/TIFS.2016.2582562",
language = "English",
volume = "11",
pages = "2453--2465",
journal = "IEEE Transactions on Information Forensics and Security",
issn = "1556-6013",
publisher = "IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC",
number = "11",

}

RIS (suitable for import to EndNote) - Download

TY - JOUR

T1 - Scaling up Class-Specific Kernel Discriminant Analysis for large-scale Face Verification

AU - Iosifidis, Alexandros

AU - Gabbouj, Moncef

PY - 2016

Y1 - 2016

N2 - In this paper, a novel approximate solution of the criterion used in non-linear class-specific discriminant subspace learning is proposed. We build on the Class-Specific Kernel Spectral Regression method which is a two-step process formed by an eigenanalysis step and a kernel regression step. Based on the structure of the intra-class and out-of-class scatter matrices, we provide a fast solution for the first step. For the second step, we propose the use of approximate kernel space definitions. We analytically show that the adoption of randomized and classspecific kernels have the effect of regularization and Nystr¨ombased approximation, respectively. We evaluate the proposed approach in face verification problems and compare it with existing approaches. Experimental results show the effectiveness and efficiency of the proposed Approximate Class-Specific Kernel Spectral Regression method, since it can provide satisfactory performance and scale well with the size of the data.

AB - In this paper, a novel approximate solution of the criterion used in non-linear class-specific discriminant subspace learning is proposed. We build on the Class-Specific Kernel Spectral Regression method which is a two-step process formed by an eigenanalysis step and a kernel regression step. Based on the structure of the intra-class and out-of-class scatter matrices, we provide a fast solution for the first step. For the second step, we propose the use of approximate kernel space definitions. We analytically show that the adoption of randomized and classspecific kernels have the effect of regularization and Nystr¨ombased approximation, respectively. We evaluate the proposed approach in face verification problems and compare it with existing approaches. Experimental results show the effectiveness and efficiency of the proposed Approximate Class-Specific Kernel Spectral Regression method, since it can provide satisfactory performance and scale well with the size of the data.

U2 - 10.1109/TIFS.2016.2582562

DO - 10.1109/TIFS.2016.2582562

M3 - Article

VL - 11

SP - 2453

EP - 2465

JO - IEEE Transactions on Information Forensics and Security

JF - IEEE Transactions on Information Forensics and Security

SN - 1556-6013

IS - 11

ER -