Video saliency detection based on random walk with restart

Jun Seong Kim, Hansang Kim, Jae Young Sim, Chang-Su Kim, Sang Uk Lee

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

1 Citation (Scopus)

Abstract

A graph-based video saliency detection algorithm is proposed in this work. We model eye movements on an image plane as random walks on a graph. To detect the saliency of the first frame in a video sequence, we construct a fully connected graph, in which each node represents an image block. We assign an edge weight to be proportional to the dissimilarity between the incident nodes and inversely proportional to their geometrical distance. We extract the saliency level of each node from the stationary distribution of the random walker on the graph. Next, to detect the saliency of each subsequent frame, we add the criterion that an edge, connecting a slow motion node to a fast motion node, should have a large weight. We then compute the stationary distribution of the random walk with restart (RWR) simulation, in which the saliency of the previous frame is used as the restarting distribution. Experimental results show that the proposed algorithm provides more reliable and accurate saliency detection performance than conventional algorithms.

Original languageEnglish
Title of host publication2013 IEEE International Conference on Image Processing, ICIP 2013 - Proceedings
Pages2465-2469
Number of pages5
DOIs
Publication statusPublished - 2013 Dec 1
Event2013 20th IEEE International Conference on Image Processing, ICIP 2013 - Melbourne, VIC, Australia
Duration: 2013 Sep 152013 Sep 18

Other

Other2013 20th IEEE International Conference on Image Processing, ICIP 2013
CountryAustralia
CityMelbourne, VIC
Period13/9/1513/9/18

Fingerprint

Eye movements

Keywords

  • Markov chain
  • random walk with restart
  • Saliency detection
  • video saliency

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition

Cite this

Kim, J. S., Kim, H., Sim, J. Y., Kim, C-S., & Lee, S. U. (2013). Video saliency detection based on random walk with restart. In 2013 IEEE International Conference on Image Processing, ICIP 2013 - Proceedings (pp. 2465-2469). [6738508] https://doi.org/10.1109/ICIP.2013.6738508

Video saliency detection based on random walk with restart. / Kim, Jun Seong; Kim, Hansang; Sim, Jae Young; Kim, Chang-Su; Lee, Sang Uk.

2013 IEEE International Conference on Image Processing, ICIP 2013 - Proceedings. 2013. p. 2465-2469 6738508.

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

Kim, JS, Kim, H, Sim, JY, Kim, C-S & Lee, SU 2013, Video saliency detection based on random walk with restart. in 2013 IEEE International Conference on Image Processing, ICIP 2013 - Proceedings., 6738508, pp. 2465-2469, 2013 20th IEEE International Conference on Image Processing, ICIP 2013, Melbourne, VIC, Australia, 13/9/15. https://doi.org/10.1109/ICIP.2013.6738508
Kim JS, Kim H, Sim JY, Kim C-S, Lee SU. Video saliency detection based on random walk with restart. In 2013 IEEE International Conference on Image Processing, ICIP 2013 - Proceedings. 2013. p. 2465-2469. 6738508 https://doi.org/10.1109/ICIP.2013.6738508
Kim, Jun Seong ; Kim, Hansang ; Sim, Jae Young ; Kim, Chang-Su ; Lee, Sang Uk. / Video saliency detection based on random walk with restart. 2013 IEEE International Conference on Image Processing, ICIP 2013 - Proceedings. 2013. pp. 2465-2469
@inproceedings{6f0a2513b8ab4703b7afa3c7fdfc8a93,
title = "Video saliency detection based on random walk with restart",
abstract = "A graph-based video saliency detection algorithm is proposed in this work. We model eye movements on an image plane as random walks on a graph. To detect the saliency of the first frame in a video sequence, we construct a fully connected graph, in which each node represents an image block. We assign an edge weight to be proportional to the dissimilarity between the incident nodes and inversely proportional to their geometrical distance. We extract the saliency level of each node from the stationary distribution of the random walker on the graph. Next, to detect the saliency of each subsequent frame, we add the criterion that an edge, connecting a slow motion node to a fast motion node, should have a large weight. We then compute the stationary distribution of the random walk with restart (RWR) simulation, in which the saliency of the previous frame is used as the restarting distribution. Experimental results show that the proposed algorithm provides more reliable and accurate saliency detection performance than conventional algorithms.",
keywords = "Markov chain, random walk with restart, Saliency detection, video saliency",
author = "Kim, {Jun Seong} and Hansang Kim and Sim, {Jae Young} and Chang-Su Kim and Lee, {Sang Uk}",
year = "2013",
month = "12",
day = "1",
doi = "10.1109/ICIP.2013.6738508",
language = "English",
isbn = "9781479923410",
pages = "2465--2469",
booktitle = "2013 IEEE International Conference on Image Processing, ICIP 2013 - Proceedings",

}

TY - GEN

T1 - Video saliency detection based on random walk with restart

AU - Kim, Jun Seong

AU - Kim, Hansang

AU - Sim, Jae Young

AU - Kim, Chang-Su

AU - Lee, Sang Uk

PY - 2013/12/1

Y1 - 2013/12/1

N2 - A graph-based video saliency detection algorithm is proposed in this work. We model eye movements on an image plane as random walks on a graph. To detect the saliency of the first frame in a video sequence, we construct a fully connected graph, in which each node represents an image block. We assign an edge weight to be proportional to the dissimilarity between the incident nodes and inversely proportional to their geometrical distance. We extract the saliency level of each node from the stationary distribution of the random walker on the graph. Next, to detect the saliency of each subsequent frame, we add the criterion that an edge, connecting a slow motion node to a fast motion node, should have a large weight. We then compute the stationary distribution of the random walk with restart (RWR) simulation, in which the saliency of the previous frame is used as the restarting distribution. Experimental results show that the proposed algorithm provides more reliable and accurate saliency detection performance than conventional algorithms.

AB - A graph-based video saliency detection algorithm is proposed in this work. We model eye movements on an image plane as random walks on a graph. To detect the saliency of the first frame in a video sequence, we construct a fully connected graph, in which each node represents an image block. We assign an edge weight to be proportional to the dissimilarity between the incident nodes and inversely proportional to their geometrical distance. We extract the saliency level of each node from the stationary distribution of the random walker on the graph. Next, to detect the saliency of each subsequent frame, we add the criterion that an edge, connecting a slow motion node to a fast motion node, should have a large weight. We then compute the stationary distribution of the random walk with restart (RWR) simulation, in which the saliency of the previous frame is used as the restarting distribution. Experimental results show that the proposed algorithm provides more reliable and accurate saliency detection performance than conventional algorithms.

KW - Markov chain

KW - random walk with restart

KW - Saliency detection

KW - video saliency

UR - http://www.scopus.com/inward/record.url?scp=84897802542&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84897802542&partnerID=8YFLogxK

U2 - 10.1109/ICIP.2013.6738508

DO - 10.1109/ICIP.2013.6738508

M3 - Conference contribution

SN - 9781479923410

SP - 2465

EP - 2469

BT - 2013 IEEE International Conference on Image Processing, ICIP 2013 - Proceedings

ER -