Video object segmentation using multiple random walkers with GMM restart rule

Minhyeok Heo, Won Dong Jang, Chang-Su Kim

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

Abstract

In this work, we propose a weakly supervised online video object segmentation algorithm, which accepts a bounding box as user annotation. First, we estimate the initial distributions of the foreground and the background by employing a visual saliency detector. Next, we simulate movements of double random walkers, one for the foreground and the other for the background. To this end, we introduce a novel restart rule based on Gaussian mixture models (GMMs). We update the GMMs during the random walk simulation to encourage interactions between the two random walkers. To achieve video segmentation, from the second to the last frames, we sequentially propagate the segmentation labels and the GMMs of the previous frame in order to maintain temporal consistency. Experimental results demonstrate that the proposed algorithm outperforms conventional video object segmentation algorithms.

Original languageEnglish
Title of host publication2016 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA 2016
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9789881476821
DOIs
Publication statusPublished - 2017 Jan 17
Event2016 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA 2016 - Jeju, Korea, Republic of
Duration: 2016 Dec 132016 Dec 16

Other

Other2016 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA 2016
CountryKorea, Republic of
CityJeju
Period16/12/1316/12/16

Fingerprint

Labels
Detectors

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Science Applications
  • Information Systems
  • Signal Processing

Cite this

Heo, M., Jang, W. D., & Kim, C-S. (2017). Video object segmentation using multiple random walkers with GMM restart rule. In 2016 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA 2016 [7820709] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/APSIPA.2016.7820709

Video object segmentation using multiple random walkers with GMM restart rule. / Heo, Minhyeok; Jang, Won Dong; Kim, Chang-Su.

2016 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA 2016. Institute of Electrical and Electronics Engineers Inc., 2017. 7820709.

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

Heo, M, Jang, WD & Kim, C-S 2017, Video object segmentation using multiple random walkers with GMM restart rule. in 2016 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA 2016., 7820709, Institute of Electrical and Electronics Engineers Inc., 2016 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA 2016, Jeju, Korea, Republic of, 16/12/13. https://doi.org/10.1109/APSIPA.2016.7820709
Heo M, Jang WD, Kim C-S. Video object segmentation using multiple random walkers with GMM restart rule. In 2016 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA 2016. Institute of Electrical and Electronics Engineers Inc. 2017. 7820709 https://doi.org/10.1109/APSIPA.2016.7820709
Heo, Minhyeok ; Jang, Won Dong ; Kim, Chang-Su. / Video object segmentation using multiple random walkers with GMM restart rule. 2016 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA 2016. Institute of Electrical and Electronics Engineers Inc., 2017.
@inproceedings{7f57a1d94f4f449c979adeabd3484e17,
title = "Video object segmentation using multiple random walkers with GMM restart rule",
abstract = "In this work, we propose a weakly supervised online video object segmentation algorithm, which accepts a bounding box as user annotation. First, we estimate the initial distributions of the foreground and the background by employing a visual saliency detector. Next, we simulate movements of double random walkers, one for the foreground and the other for the background. To this end, we introduce a novel restart rule based on Gaussian mixture models (GMMs). We update the GMMs during the random walk simulation to encourage interactions between the two random walkers. To achieve video segmentation, from the second to the last frames, we sequentially propagate the segmentation labels and the GMMs of the previous frame in order to maintain temporal consistency. Experimental results demonstrate that the proposed algorithm outperforms conventional video object segmentation algorithms.",
author = "Minhyeok Heo and Jang, {Won Dong} and Chang-Su Kim",
year = "2017",
month = "1",
day = "17",
doi = "10.1109/APSIPA.2016.7820709",
language = "English",
booktitle = "2016 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA 2016",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - Video object segmentation using multiple random walkers with GMM restart rule

AU - Heo, Minhyeok

AU - Jang, Won Dong

AU - Kim, Chang-Su

PY - 2017/1/17

Y1 - 2017/1/17

N2 - In this work, we propose a weakly supervised online video object segmentation algorithm, which accepts a bounding box as user annotation. First, we estimate the initial distributions of the foreground and the background by employing a visual saliency detector. Next, we simulate movements of double random walkers, one for the foreground and the other for the background. To this end, we introduce a novel restart rule based on Gaussian mixture models (GMMs). We update the GMMs during the random walk simulation to encourage interactions between the two random walkers. To achieve video segmentation, from the second to the last frames, we sequentially propagate the segmentation labels and the GMMs of the previous frame in order to maintain temporal consistency. Experimental results demonstrate that the proposed algorithm outperforms conventional video object segmentation algorithms.

AB - In this work, we propose a weakly supervised online video object segmentation algorithm, which accepts a bounding box as user annotation. First, we estimate the initial distributions of the foreground and the background by employing a visual saliency detector. Next, we simulate movements of double random walkers, one for the foreground and the other for the background. To this end, we introduce a novel restart rule based on Gaussian mixture models (GMMs). We update the GMMs during the random walk simulation to encourage interactions between the two random walkers. To achieve video segmentation, from the second to the last frames, we sequentially propagate the segmentation labels and the GMMs of the previous frame in order to maintain temporal consistency. Experimental results demonstrate that the proposed algorithm outperforms conventional video object segmentation algorithms.

UR - http://www.scopus.com/inward/record.url?scp=85013848051&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85013848051&partnerID=8YFLogxK

U2 - 10.1109/APSIPA.2016.7820709

DO - 10.1109/APSIPA.2016.7820709

M3 - Conference contribution

AN - SCOPUS:85013848051

BT - 2016 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA 2016

PB - Institute of Electrical and Electronics Engineers Inc.

ER -