Constant-time bilateral filter using spectral decomposition

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

6 Citations (Scopus)

Abstract

This paper presents an efficient constant-time bilateral filter where constant-time means that computational complexity is independent of filter window size. Many state-of-the-art constant-time methods approximate the original bilateral filter by an appropriate combination of a series of convolutions. It is important for this framework to optimize the performance tradeoff between approximate accuracy and the number of convolutions. The proposed method achieves the optimal performance tradeoff in a least-squares manner by using spectral decomposition under the assumption that images consist of discrete intensities such as 8-bit images. This approach is essentially applicable to arbitrary range kernel. Experiments show that the proposed method outperforms state-of-the-art methods in terms of both computational complexity and approximate accuracy.

Original languageEnglish
Title of host publication2016 IEEE International Conference on Image Processing, ICIP 2016 - Proceedings
PublisherIEEE Computer Society
Pages3319-3323
Number of pages5
Volume2016-August
ISBN (Electronic)9781467399616
DOIs
Publication statusPublished - 2016 Aug 3
Event23rd IEEE International Conference on Image Processing, ICIP 2016 - Phoenix, United States
Duration: 2016 Sep 252016 Sep 28

Other

Other23rd IEEE International Conference on Image Processing, ICIP 2016
CountryUnited States
CityPhoenix
Period16/9/2516/9/28

Fingerprint

Convolution
Computational complexity
Decomposition
Experiments

Keywords

  • Constant-time bilateral filter
  • Image filtering
  • Spectral decomposition

ASJC Scopus subject areas

  • Software
  • Computer Vision and Pattern Recognition
  • Signal Processing

Cite this

Sugimoto, K., Breckon, T., & Kamata, S. (2016). Constant-time bilateral filter using spectral decomposition. In 2016 IEEE International Conference on Image Processing, ICIP 2016 - Proceedings (Vol. 2016-August, pp. 3319-3323). [7532974] IEEE Computer Society. https://doi.org/10.1109/ICIP.2016.7532974

Constant-time bilateral filter using spectral decomposition. / Sugimoto, Kenjiro; Breckon, Toby; Kamata, Seiichiro.

2016 IEEE International Conference on Image Processing, ICIP 2016 - Proceedings. Vol. 2016-August IEEE Computer Society, 2016. p. 3319-3323 7532974.

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

Sugimoto, K, Breckon, T & Kamata, S 2016, Constant-time bilateral filter using spectral decomposition. in 2016 IEEE International Conference on Image Processing, ICIP 2016 - Proceedings. vol. 2016-August, 7532974, IEEE Computer Society, pp. 3319-3323, 23rd IEEE International Conference on Image Processing, ICIP 2016, Phoenix, United States, 16/9/25. https://doi.org/10.1109/ICIP.2016.7532974
Sugimoto K, Breckon T, Kamata S. Constant-time bilateral filter using spectral decomposition. In 2016 IEEE International Conference on Image Processing, ICIP 2016 - Proceedings. Vol. 2016-August. IEEE Computer Society. 2016. p. 3319-3323. 7532974 https://doi.org/10.1109/ICIP.2016.7532974
Sugimoto, Kenjiro ; Breckon, Toby ; Kamata, Seiichiro. / Constant-time bilateral filter using spectral decomposition. 2016 IEEE International Conference on Image Processing, ICIP 2016 - Proceedings. Vol. 2016-August IEEE Computer Society, 2016. pp. 3319-3323
@inproceedings{39e089218a284fec990b18173be74e1f,
title = "Constant-time bilateral filter using spectral decomposition",
abstract = "This paper presents an efficient constant-time bilateral filter where constant-time means that computational complexity is independent of filter window size. Many state-of-the-art constant-time methods approximate the original bilateral filter by an appropriate combination of a series of convolutions. It is important for this framework to optimize the performance tradeoff between approximate accuracy and the number of convolutions. The proposed method achieves the optimal performance tradeoff in a least-squares manner by using spectral decomposition under the assumption that images consist of discrete intensities such as 8-bit images. This approach is essentially applicable to arbitrary range kernel. Experiments show that the proposed method outperforms state-of-the-art methods in terms of both computational complexity and approximate accuracy.",
keywords = "Constant-time bilateral filter, Image filtering, Spectral decomposition",
author = "Kenjiro Sugimoto and Toby Breckon and Seiichiro Kamata",
year = "2016",
month = "8",
day = "3",
doi = "10.1109/ICIP.2016.7532974",
language = "English",
volume = "2016-August",
pages = "3319--3323",
booktitle = "2016 IEEE International Conference on Image Processing, ICIP 2016 - Proceedings",
publisher = "IEEE Computer Society",
address = "United States",

}

TY - GEN

T1 - Constant-time bilateral filter using spectral decomposition

AU - Sugimoto, Kenjiro

AU - Breckon, Toby

AU - Kamata, Seiichiro

PY - 2016/8/3

Y1 - 2016/8/3

N2 - This paper presents an efficient constant-time bilateral filter where constant-time means that computational complexity is independent of filter window size. Many state-of-the-art constant-time methods approximate the original bilateral filter by an appropriate combination of a series of convolutions. It is important for this framework to optimize the performance tradeoff between approximate accuracy and the number of convolutions. The proposed method achieves the optimal performance tradeoff in a least-squares manner by using spectral decomposition under the assumption that images consist of discrete intensities such as 8-bit images. This approach is essentially applicable to arbitrary range kernel. Experiments show that the proposed method outperforms state-of-the-art methods in terms of both computational complexity and approximate accuracy.

AB - This paper presents an efficient constant-time bilateral filter where constant-time means that computational complexity is independent of filter window size. Many state-of-the-art constant-time methods approximate the original bilateral filter by an appropriate combination of a series of convolutions. It is important for this framework to optimize the performance tradeoff between approximate accuracy and the number of convolutions. The proposed method achieves the optimal performance tradeoff in a least-squares manner by using spectral decomposition under the assumption that images consist of discrete intensities such as 8-bit images. This approach is essentially applicable to arbitrary range kernel. Experiments show that the proposed method outperforms state-of-the-art methods in terms of both computational complexity and approximate accuracy.

KW - Constant-time bilateral filter

KW - Image filtering

KW - Spectral decomposition

UR - http://www.scopus.com/inward/record.url?scp=85006795561&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85006795561&partnerID=8YFLogxK

U2 - 10.1109/ICIP.2016.7532974

DO - 10.1109/ICIP.2016.7532974

M3 - Conference contribution

AN - SCOPUS:85006795561

VL - 2016-August

SP - 3319

EP - 3323

BT - 2016 IEEE International Conference on Image Processing, ICIP 2016 - Proceedings

PB - IEEE Computer Society

ER -