Home > Research > Publications & Outputs > Local high-order regularization on data manifolds

Electronic data

  • HighorderRegularization

    Rights statement: ©2015 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, 3.4 MB, PDF document

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

Links

Text available via DOI:

View graph of relations

Local high-order regularization on data manifolds

Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSNConference contribution/Paperpeer-review

Published

Standard

Local high-order regularization on data manifolds. / Kim, Kwang In; Tompkin, James; Pfister, Hanspeter ; Theobalt, Christian.

Computer Vision and Pattern Recognition (CVPR), 2015 IEEE Conference on . IEEE, 2015. p. 5473-5481.

Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSNConference contribution/Paperpeer-review

Harvard

Kim, KI, Tompkin, J, Pfister, H & Theobalt, C 2015, Local high-order regularization on data manifolds. in Computer Vision and Pattern Recognition (CVPR), 2015 IEEE Conference on . IEEE, pp. 5473-5481. https://doi.org/10.1109/CVPR.2015.7299186

APA

Kim, K. I., Tompkin, J., Pfister, H., & Theobalt, C. (2015). Local high-order regularization on data manifolds. In Computer Vision and Pattern Recognition (CVPR), 2015 IEEE Conference on (pp. 5473-5481). IEEE. https://doi.org/10.1109/CVPR.2015.7299186

Vancouver

Kim KI, Tompkin J, Pfister H, Theobalt C. Local high-order regularization on data manifolds. In Computer Vision and Pattern Recognition (CVPR), 2015 IEEE Conference on . IEEE. 2015. p. 5473-5481 https://doi.org/10.1109/CVPR.2015.7299186

Author

Kim, Kwang In ; Tompkin, James ; Pfister, Hanspeter ; Theobalt, Christian. / Local high-order regularization on data manifolds. Computer Vision and Pattern Recognition (CVPR), 2015 IEEE Conference on . IEEE, 2015. pp. 5473-5481

Bibtex

@inproceedings{2e0983693b9749408aa9d4186827b411,
title = "Local high-order regularization on data manifolds",
abstract = "The common graph Laplacian regularizer is well-established in semi-supervised learning and spectral dimensionality reduction. However, as a first-order regularizer, it can lead to degenerate functions in high-dimensional manifolds. The iterated graph Laplacian enables high-order regularization, but it has a high computational complexity and so cannot be applied to large problems. We introduce a new regularizer which is globally high order and so does not suffer from the degeneracy of the graph Laplacian regularizer, but is also sparse for efficient computation in semi-supervised learning applications. We reduce computational complexity by building a local first-order approximation of the manifold as a surrogate geometry, and construct our high-order regularizer based on local derivative evaluations therein. Experiments on human body shape and pose analysis demonstrate the effectiveness and efficiency of our method.",
author = "Kim, {Kwang In} and James Tompkin and Hanspeter Pfister and Christian Theobalt",
note = " {\textcopyright}2015 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 = "2015",
month = jun,
day = "8",
doi = "10.1109/CVPR.2015.7299186",
language = "English",
pages = "5473--5481",
booktitle = "Computer Vision and Pattern Recognition (CVPR), 2015 IEEE Conference on",
publisher = "IEEE",

}

RIS

TY - GEN

T1 - Local high-order regularization on data manifolds

AU - Kim, Kwang In

AU - Tompkin, James

AU - Pfister, Hanspeter

AU - Theobalt, Christian

N1 - ©2015 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 - 2015/6/8

Y1 - 2015/6/8

N2 - The common graph Laplacian regularizer is well-established in semi-supervised learning and spectral dimensionality reduction. However, as a first-order regularizer, it can lead to degenerate functions in high-dimensional manifolds. The iterated graph Laplacian enables high-order regularization, but it has a high computational complexity and so cannot be applied to large problems. We introduce a new regularizer which is globally high order and so does not suffer from the degeneracy of the graph Laplacian regularizer, but is also sparse for efficient computation in semi-supervised learning applications. We reduce computational complexity by building a local first-order approximation of the manifold as a surrogate geometry, and construct our high-order regularizer based on local derivative evaluations therein. Experiments on human body shape and pose analysis demonstrate the effectiveness and efficiency of our method.

AB - The common graph Laplacian regularizer is well-established in semi-supervised learning and spectral dimensionality reduction. However, as a first-order regularizer, it can lead to degenerate functions in high-dimensional manifolds. The iterated graph Laplacian enables high-order regularization, but it has a high computational complexity and so cannot be applied to large problems. We introduce a new regularizer which is globally high order and so does not suffer from the degeneracy of the graph Laplacian regularizer, but is also sparse for efficient computation in semi-supervised learning applications. We reduce computational complexity by building a local first-order approximation of the manifold as a surrogate geometry, and construct our high-order regularizer based on local derivative evaluations therein. Experiments on human body shape and pose analysis demonstrate the effectiveness and efficiency of our method.

U2 - 10.1109/CVPR.2015.7299186

DO - 10.1109/CVPR.2015.7299186

M3 - Conference contribution/Paper

SP - 5473

EP - 5481

BT - Computer Vision and Pattern Recognition (CVPR), 2015 IEEE Conference on

PB - IEEE

ER -