Home > Research > Publications & Outputs > Iterative kernel principal component analysis f...
View graph of relations

Iterative kernel principal component analysis for image modeling

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Published

Standard

Iterative kernel principal component analysis for image modeling. / Kim, Kwang In; Franz, Matthias O.; Schölkopf, Bernhard.
In: IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 27, No. 9, 2005, p. 1351-1366.

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Harvard

Kim, KI, Franz, MO & Schölkopf, B 2005, 'Iterative kernel principal component analysis for image modeling', IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 27, no. 9, pp. 1351-1366. https://doi.org/10.1109/TPAMI.2005.181

APA

Kim, K. I., Franz, M. O., & Schölkopf, B. (2005). Iterative kernel principal component analysis for image modeling. IEEE Transactions on Pattern Analysis and Machine Intelligence, 27(9), 1351-1366. https://doi.org/10.1109/TPAMI.2005.181

Vancouver

Kim KI, Franz MO, Schölkopf B. Iterative kernel principal component analysis for image modeling. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2005;27(9):1351-1366. doi: 10.1109/TPAMI.2005.181

Author

Kim, Kwang In ; Franz, Matthias O. ; Schölkopf, Bernhard. / Iterative kernel principal component analysis for image modeling. In: IEEE Transactions on Pattern Analysis and Machine Intelligence. 2005 ; Vol. 27, No. 9. pp. 1351-1366.

Bibtex

@article{3f478f8e7dfc440fa7dc1f0ffe208db4,
title = "Iterative kernel principal component analysis for image modeling",
abstract = "In recent years, Kernel Principal Component Analysis (KPCA) has been suggested for various image processing tasks requiring an image model such as, e.g., denoising or compression. The original form of KPCA, however, can be only applied to strongly restricted image classes due to the limited number of training examples that can be processed. We therefore propose a new iterative method for performing KPCA, the Kernel Hebbian Algorithm which iteratively estimates the Kernel Principal Components with only linear order memory complexity. In our experiments, we compute models for complex image classes such as faces and natural images which require a large number of training examples. The resulting image models are tested in single-frame super-resolution and denoising applications. The KPCA model is not specifically tailored to these tasks; in fact, the same model can be used in super-resolution with variable input resolution, or denoising with unknown noise characteristics. In spite of this, both super-resolution and denoising performance are comparable to existing methods.",
author = "Kim, {Kwang In} and Franz, {Matthias O.} and Bernhard Sch{\"o}lkopf",
year = "2005",
doi = "10.1109/TPAMI.2005.181",
language = "English",
volume = "27",
pages = "1351--1366",
journal = "IEEE Transactions on Pattern Analysis and Machine Intelligence",
issn = "0162-8828",
publisher = "IEEE Computer Society",
number = "9",

}

RIS

TY - JOUR

T1 - Iterative kernel principal component analysis for image modeling

AU - Kim, Kwang In

AU - Franz, Matthias O.

AU - Schölkopf, Bernhard

PY - 2005

Y1 - 2005

N2 - In recent years, Kernel Principal Component Analysis (KPCA) has been suggested for various image processing tasks requiring an image model such as, e.g., denoising or compression. The original form of KPCA, however, can be only applied to strongly restricted image classes due to the limited number of training examples that can be processed. We therefore propose a new iterative method for performing KPCA, the Kernel Hebbian Algorithm which iteratively estimates the Kernel Principal Components with only linear order memory complexity. In our experiments, we compute models for complex image classes such as faces and natural images which require a large number of training examples. The resulting image models are tested in single-frame super-resolution and denoising applications. The KPCA model is not specifically tailored to these tasks; in fact, the same model can be used in super-resolution with variable input resolution, or denoising with unknown noise characteristics. In spite of this, both super-resolution and denoising performance are comparable to existing methods.

AB - In recent years, Kernel Principal Component Analysis (KPCA) has been suggested for various image processing tasks requiring an image model such as, e.g., denoising or compression. The original form of KPCA, however, can be only applied to strongly restricted image classes due to the limited number of training examples that can be processed. We therefore propose a new iterative method for performing KPCA, the Kernel Hebbian Algorithm which iteratively estimates the Kernel Principal Components with only linear order memory complexity. In our experiments, we compute models for complex image classes such as faces and natural images which require a large number of training examples. The resulting image models are tested in single-frame super-resolution and denoising applications. The KPCA model is not specifically tailored to these tasks; in fact, the same model can be used in super-resolution with variable input resolution, or denoising with unknown noise characteristics. In spite of this, both super-resolution and denoising performance are comparable to existing methods.

U2 - 10.1109/TPAMI.2005.181

DO - 10.1109/TPAMI.2005.181

M3 - Journal article

VL - 27

SP - 1351

EP - 1366

JO - IEEE Transactions on Pattern Analysis and Machine Intelligence

JF - IEEE Transactions on Pattern Analysis and Machine Intelligence

SN - 0162-8828

IS - 9

ER -