Edge-aware image smoothing using commute time distances

Youngjung Kim, Changjae Oh, Kwanghoon Sohn

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

4 Citations (Scopus)

Abstract

Most edge-aware smoothing methods are based on the Euclidean distance to measure the similarity between adjacent pixels. This paper exploits the properties of the commute time to extend the notion of 'similarity' in this context. The intuition is that since the commute time reflects the effect of all possible weighted paths between nodes (pixels), it can account for the global distribution of image features. The commute time is characterized by eigenvectors of a large Laplacian matrix, which is very costly even with sophisticated eigen-solver. To this end, we further employ a multiscale algorithm for approximating the eigenvector computation efficiently. It is analogous to the classical Nystrom's method for low rank matrix approximation. However, we do not depend on long-range connections between nodes, allowing one to include spatial coordinates in defining feature space. Extensive experimental validation demonstrates the benefits of using the commute time in a range of image processing applications, such as edge-aware image smoothing, texture filtering, and local edit propagation.

Original languageEnglish
Title of host publication2016 IEEE International Conference on Image Processing, ICIP 2016 - Proceedings
PublisherIEEE Computer Society
Pages3304-3308
Number of pages5
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

Publication series

NameProceedings - International Conference on Image Processing, ICIP
Volume2016-August
ISSN (Print)1522-4880

Other

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

Fingerprint

Eigenvalues and eigenfunctions
Pixels
Image texture
Image processing

All Science Journal Classification (ASJC) codes

  • Software
  • Computer Vision and Pattern Recognition
  • Signal Processing

Cite this

Kim, Y., Oh, C., & Sohn, K. (2016). Edge-aware image smoothing using commute time distances. In 2016 IEEE International Conference on Image Processing, ICIP 2016 - Proceedings (pp. 3304-3308). [7532971] (Proceedings - International Conference on Image Processing, ICIP; Vol. 2016-August). IEEE Computer Society. https://doi.org/10.1109/ICIP.2016.7532971
Kim, Youngjung ; Oh, Changjae ; Sohn, Kwanghoon. / Edge-aware image smoothing using commute time distances. 2016 IEEE International Conference on Image Processing, ICIP 2016 - Proceedings. IEEE Computer Society, 2016. pp. 3304-3308 (Proceedings - International Conference on Image Processing, ICIP).
@inproceedings{d4fef34cbc3740f3914e99a1445705ad,
title = "Edge-aware image smoothing using commute time distances",
abstract = "Most edge-aware smoothing methods are based on the Euclidean distance to measure the similarity between adjacent pixels. This paper exploits the properties of the commute time to extend the notion of 'similarity' in this context. The intuition is that since the commute time reflects the effect of all possible weighted paths between nodes (pixels), it can account for the global distribution of image features. The commute time is characterized by eigenvectors of a large Laplacian matrix, which is very costly even with sophisticated eigen-solver. To this end, we further employ a multiscale algorithm for approximating the eigenvector computation efficiently. It is analogous to the classical Nystrom's method for low rank matrix approximation. However, we do not depend on long-range connections between nodes, allowing one to include spatial coordinates in defining feature space. Extensive experimental validation demonstrates the benefits of using the commute time in a range of image processing applications, such as edge-aware image smoothing, texture filtering, and local edit propagation.",
author = "Youngjung Kim and Changjae Oh and Kwanghoon Sohn",
year = "2016",
month = "8",
day = "3",
doi = "10.1109/ICIP.2016.7532971",
language = "English",
series = "Proceedings - International Conference on Image Processing, ICIP",
publisher = "IEEE Computer Society",
pages = "3304--3308",
booktitle = "2016 IEEE International Conference on Image Processing, ICIP 2016 - Proceedings",
address = "United States",

}

Kim, Y, Oh, C & Sohn, K 2016, Edge-aware image smoothing using commute time distances. in 2016 IEEE International Conference on Image Processing, ICIP 2016 - Proceedings., 7532971, Proceedings - International Conference on Image Processing, ICIP, vol. 2016-August, IEEE Computer Society, pp. 3304-3308, 23rd IEEE International Conference on Image Processing, ICIP 2016, Phoenix, United States, 16/9/25. https://doi.org/10.1109/ICIP.2016.7532971

Edge-aware image smoothing using commute time distances. / Kim, Youngjung; Oh, Changjae; Sohn, Kwanghoon.

2016 IEEE International Conference on Image Processing, ICIP 2016 - Proceedings. IEEE Computer Society, 2016. p. 3304-3308 7532971 (Proceedings - International Conference on Image Processing, ICIP; Vol. 2016-August).

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

TY - GEN

T1 - Edge-aware image smoothing using commute time distances

AU - Kim, Youngjung

AU - Oh, Changjae

AU - Sohn, Kwanghoon

PY - 2016/8/3

Y1 - 2016/8/3

N2 - Most edge-aware smoothing methods are based on the Euclidean distance to measure the similarity between adjacent pixels. This paper exploits the properties of the commute time to extend the notion of 'similarity' in this context. The intuition is that since the commute time reflects the effect of all possible weighted paths between nodes (pixels), it can account for the global distribution of image features. The commute time is characterized by eigenvectors of a large Laplacian matrix, which is very costly even with sophisticated eigen-solver. To this end, we further employ a multiscale algorithm for approximating the eigenvector computation efficiently. It is analogous to the classical Nystrom's method for low rank matrix approximation. However, we do not depend on long-range connections between nodes, allowing one to include spatial coordinates in defining feature space. Extensive experimental validation demonstrates the benefits of using the commute time in a range of image processing applications, such as edge-aware image smoothing, texture filtering, and local edit propagation.

AB - Most edge-aware smoothing methods are based on the Euclidean distance to measure the similarity between adjacent pixels. This paper exploits the properties of the commute time to extend the notion of 'similarity' in this context. The intuition is that since the commute time reflects the effect of all possible weighted paths between nodes (pixels), it can account for the global distribution of image features. The commute time is characterized by eigenvectors of a large Laplacian matrix, which is very costly even with sophisticated eigen-solver. To this end, we further employ a multiscale algorithm for approximating the eigenvector computation efficiently. It is analogous to the classical Nystrom's method for low rank matrix approximation. However, we do not depend on long-range connections between nodes, allowing one to include spatial coordinates in defining feature space. Extensive experimental validation demonstrates the benefits of using the commute time in a range of image processing applications, such as edge-aware image smoothing, texture filtering, and local edit propagation.

UR - http://www.scopus.com/inward/record.url?scp=85006789776&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85006789776&partnerID=8YFLogxK

U2 - 10.1109/ICIP.2016.7532971

DO - 10.1109/ICIP.2016.7532971

M3 - Conference contribution

T3 - Proceedings - International Conference on Image Processing, ICIP

SP - 3304

EP - 3308

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

PB - IEEE Computer Society

ER -

Kim Y, Oh C, Sohn K. Edge-aware image smoothing using commute time distances. In 2016 IEEE International Conference on Image Processing, ICIP 2016 - Proceedings. IEEE Computer Society. 2016. p. 3304-3308. 7532971. (Proceedings - International Conference on Image Processing, ICIP). https://doi.org/10.1109/ICIP.2016.7532971