Bilateral tangent projection for manifold smoothing

Yi Hu, Lei Lei, Won Jae Park, Sung-Jea Ko

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In this paper, we propose a bilateral tangent projection (BTP) algorithm to smooth a set of noisy data assumed to lie on a low-dimensional manifold. The data points are smoothed by a family of local tangent projections adaptively. The BTP algorithm utilizes not only the location information of the data point but also the fitness of the tangent space estimation. The local tangent space with better fitness plays more important role in the smoothing procedure. The experimental results show that our algorithm is effective for noise reduction and preserves the structure of the underlying manifold.

Original languageEnglish
Title of host publicationProceedings - 4th International Congress on Image and Signal Processing, CISP 2011
Pages2210-2213
Number of pages4
Volume4
DOIs
Publication statusPublished - 2011 Dec 1
Event4th International Congress on Image and Signal Processing, CISP 2011 - Shanghai, China
Duration: 2011 Oct 152011 Oct 17

Other

Other4th International Congress on Image and Signal Processing, CISP 2011
CountryChina
CityShanghai
Period11/10/1511/10/17

Fingerprint

Noise abatement

Keywords

  • denoising
  • manifold learning
  • manifold smoothing

ASJC Scopus subject areas

  • Computer Graphics and Computer-Aided Design
  • Computer Vision and Pattern Recognition
  • Signal Processing

Cite this

Hu, Y., Lei, L., Park, W. J., & Ko, S-J. (2011). Bilateral tangent projection for manifold smoothing. In Proceedings - 4th International Congress on Image and Signal Processing, CISP 2011 (Vol. 4, pp. 2210-2213). [6100601] https://doi.org/10.1109/CISP.2011.6100601

Bilateral tangent projection for manifold smoothing. / Hu, Yi; Lei, Lei; Park, Won Jae; Ko, Sung-Jea.

Proceedings - 4th International Congress on Image and Signal Processing, CISP 2011. Vol. 4 2011. p. 2210-2213 6100601.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Hu, Y, Lei, L, Park, WJ & Ko, S-J 2011, Bilateral tangent projection for manifold smoothing. in Proceedings - 4th International Congress on Image and Signal Processing, CISP 2011. vol. 4, 6100601, pp. 2210-2213, 4th International Congress on Image and Signal Processing, CISP 2011, Shanghai, China, 11/10/15. https://doi.org/10.1109/CISP.2011.6100601
Hu Y, Lei L, Park WJ, Ko S-J. Bilateral tangent projection for manifold smoothing. In Proceedings - 4th International Congress on Image and Signal Processing, CISP 2011. Vol. 4. 2011. p. 2210-2213. 6100601 https://doi.org/10.1109/CISP.2011.6100601
Hu, Yi ; Lei, Lei ; Park, Won Jae ; Ko, Sung-Jea. / Bilateral tangent projection for manifold smoothing. Proceedings - 4th International Congress on Image and Signal Processing, CISP 2011. Vol. 4 2011. pp. 2210-2213
@inproceedings{2eb607ebe4a349dabd56d0598bc98cf5,
title = "Bilateral tangent projection for manifold smoothing",
abstract = "In this paper, we propose a bilateral tangent projection (BTP) algorithm to smooth a set of noisy data assumed to lie on a low-dimensional manifold. The data points are smoothed by a family of local tangent projections adaptively. The BTP algorithm utilizes not only the location information of the data point but also the fitness of the tangent space estimation. The local tangent space with better fitness plays more important role in the smoothing procedure. The experimental results show that our algorithm is effective for noise reduction and preserves the structure of the underlying manifold.",
keywords = "denoising, manifold learning, manifold smoothing",
author = "Yi Hu and Lei Lei and Park, {Won Jae} and Sung-Jea Ko",
year = "2011",
month = "12",
day = "1",
doi = "10.1109/CISP.2011.6100601",
language = "English",
isbn = "9781424493067",
volume = "4",
pages = "2210--2213",
booktitle = "Proceedings - 4th International Congress on Image and Signal Processing, CISP 2011",

}

TY - GEN

T1 - Bilateral tangent projection for manifold smoothing

AU - Hu, Yi

AU - Lei, Lei

AU - Park, Won Jae

AU - Ko, Sung-Jea

PY - 2011/12/1

Y1 - 2011/12/1

N2 - In this paper, we propose a bilateral tangent projection (BTP) algorithm to smooth a set of noisy data assumed to lie on a low-dimensional manifold. The data points are smoothed by a family of local tangent projections adaptively. The BTP algorithm utilizes not only the location information of the data point but also the fitness of the tangent space estimation. The local tangent space with better fitness plays more important role in the smoothing procedure. The experimental results show that our algorithm is effective for noise reduction and preserves the structure of the underlying manifold.

AB - In this paper, we propose a bilateral tangent projection (BTP) algorithm to smooth a set of noisy data assumed to lie on a low-dimensional manifold. The data points are smoothed by a family of local tangent projections adaptively. The BTP algorithm utilizes not only the location information of the data point but also the fitness of the tangent space estimation. The local tangent space with better fitness plays more important role in the smoothing procedure. The experimental results show that our algorithm is effective for noise reduction and preserves the structure of the underlying manifold.

KW - denoising

KW - manifold learning

KW - manifold smoothing

UR - http://www.scopus.com/inward/record.url?scp=84862968856&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84862968856&partnerID=8YFLogxK

U2 - 10.1109/CISP.2011.6100601

DO - 10.1109/CISP.2011.6100601

M3 - Conference contribution

SN - 9781424493067

VL - 4

SP - 2210

EP - 2213

BT - Proceedings - 4th International Congress on Image and Signal Processing, CISP 2011

ER -