Real-time segmentation of objects from video sequences with non-stationary backgrounds using spatio-temporal coherence

Jae Kyun Ahn, Chang-Su Kim

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

10 Citations (Scopus)

Abstract

A real-time video segmentation algorithm, which can extract objects from video sequences even with non-stationary backgrounds, is proposed in this work. First, we segment the first frame into an object and a background interactively to build the probability density functions of colors in the object and the background. Then, for each subsequent frame, we construct a coherence strip, which is likely to contain the object contour, by exploiting spatio-temporal correlations. Finally, we perform the segmentation by minimizing an energy function composed of color, coherence, and smoothness terms. Experimental results on various test sequences show that the proposed algorithm provides accurate segmentation results in real-time, even though video sequences contain unstable camera motions.

Original languageEnglish
Title of host publicationProceedings - International Conference on Image Processing, ICIP
Pages1544-1547
Number of pages4
DOIs
Publication statusPublished - 2008 Dec 1
Event2008 IEEE International Conference on Image Processing, ICIP 2008 - San Diego, CA, United States
Duration: 2008 Oct 122008 Oct 15

Other

Other2008 IEEE International Conference on Image Processing, ICIP 2008
CountryUnited States
CitySan Diego, CA
Period08/10/1208/10/15

Fingerprint

Color
Probability density function
Cameras

Keywords

  • Graph cut
  • Kernel density estimation
  • Segmentation
  • Spatio-temporal coherence
  • Video object

ASJC Scopus subject areas

  • Software
  • Computer Vision and Pattern Recognition
  • Signal Processing

Cite this

Ahn, J. K., & Kim, C-S. (2008). Real-time segmentation of objects from video sequences with non-stationary backgrounds using spatio-temporal coherence. In Proceedings - International Conference on Image Processing, ICIP (pp. 1544-1547). [4712062] https://doi.org/10.1109/ICIP.2008.4712062

Real-time segmentation of objects from video sequences with non-stationary backgrounds using spatio-temporal coherence. / Ahn, Jae Kyun; Kim, Chang-Su.

Proceedings - International Conference on Image Processing, ICIP. 2008. p. 1544-1547 4712062.

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

Ahn, JK & Kim, C-S 2008, Real-time segmentation of objects from video sequences with non-stationary backgrounds using spatio-temporal coherence. in Proceedings - International Conference on Image Processing, ICIP., 4712062, pp. 1544-1547, 2008 IEEE International Conference on Image Processing, ICIP 2008, San Diego, CA, United States, 08/10/12. https://doi.org/10.1109/ICIP.2008.4712062
Ahn, Jae Kyun ; Kim, Chang-Su. / Real-time segmentation of objects from video sequences with non-stationary backgrounds using spatio-temporal coherence. Proceedings - International Conference on Image Processing, ICIP. 2008. pp. 1544-1547
@inproceedings{52b23d8aa2f741f3aebfcd5621575a0d,
title = "Real-time segmentation of objects from video sequences with non-stationary backgrounds using spatio-temporal coherence",
abstract = "A real-time video segmentation algorithm, which can extract objects from video sequences even with non-stationary backgrounds, is proposed in this work. First, we segment the first frame into an object and a background interactively to build the probability density functions of colors in the object and the background. Then, for each subsequent frame, we construct a coherence strip, which is likely to contain the object contour, by exploiting spatio-temporal correlations. Finally, we perform the segmentation by minimizing an energy function composed of color, coherence, and smoothness terms. Experimental results on various test sequences show that the proposed algorithm provides accurate segmentation results in real-time, even though video sequences contain unstable camera motions.",
keywords = "Graph cut, Kernel density estimation, Segmentation, Spatio-temporal coherence, Video object",
author = "Ahn, {Jae Kyun} and Chang-Su Kim",
year = "2008",
month = "12",
day = "1",
doi = "10.1109/ICIP.2008.4712062",
language = "English",
isbn = "1424417643",
pages = "1544--1547",
booktitle = "Proceedings - International Conference on Image Processing, ICIP",

}

TY - GEN

T1 - Real-time segmentation of objects from video sequences with non-stationary backgrounds using spatio-temporal coherence

AU - Ahn, Jae Kyun

AU - Kim, Chang-Su

PY - 2008/12/1

Y1 - 2008/12/1

N2 - A real-time video segmentation algorithm, which can extract objects from video sequences even with non-stationary backgrounds, is proposed in this work. First, we segment the first frame into an object and a background interactively to build the probability density functions of colors in the object and the background. Then, for each subsequent frame, we construct a coherence strip, which is likely to contain the object contour, by exploiting spatio-temporal correlations. Finally, we perform the segmentation by minimizing an energy function composed of color, coherence, and smoothness terms. Experimental results on various test sequences show that the proposed algorithm provides accurate segmentation results in real-time, even though video sequences contain unstable camera motions.

AB - A real-time video segmentation algorithm, which can extract objects from video sequences even with non-stationary backgrounds, is proposed in this work. First, we segment the first frame into an object and a background interactively to build the probability density functions of colors in the object and the background. Then, for each subsequent frame, we construct a coherence strip, which is likely to contain the object contour, by exploiting spatio-temporal correlations. Finally, we perform the segmentation by minimizing an energy function composed of color, coherence, and smoothness terms. Experimental results on various test sequences show that the proposed algorithm provides accurate segmentation results in real-time, even though video sequences contain unstable camera motions.

KW - Graph cut

KW - Kernel density estimation

KW - Segmentation

KW - Spatio-temporal coherence

KW - Video object

UR - http://www.scopus.com/inward/record.url?scp=69949144745&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=69949144745&partnerID=8YFLogxK

U2 - 10.1109/ICIP.2008.4712062

DO - 10.1109/ICIP.2008.4712062

M3 - Conference contribution

AN - SCOPUS:69949144745

SN - 1424417643

SN - 9781424417643

SP - 1544

EP - 1547

BT - Proceedings - International Conference on Image Processing, ICIP

ER -