High-quality 2D to 3D video conversion based on robust MRF-based object tracking and reliable graph-cut-based contour refinement

Joonsoo Kim, Yoonsik Choe, Yongoo Kim

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

2 Citations (Scopus)

Abstract

Generation of object-boundary-matched depth map is important to semi-automatic 2D to 3D conversion for high quality perception of converted 3D video. This paper proposes an accurate depth map generation method using an MRF-based initial contour tracking followed by a robust graph-cut-based contour refinement. The proposed MRF-based contour tracking puts a constraint on the variation of angular-radial distance for robust modeling of object motion in the presence of shape deformation. In addition, the proposed contour refinement eliminates ambiguous seeds, which might cause confusion in graph-cut segmentation, resulting in accurate contour in the presence of complex background. Experiment results show the superior tracking performance of the proposed scheme, especially when there is complex background and/or object deformation, indicating the power of the proposed scheme in object-boundary- matched accurate depth map generation.

Original languageEnglish
Title of host publication2011 International Conference on ICT Convergence, ICTC 2011
Pages360-365
Number of pages6
DOIs
Publication statusPublished - 2011 Dec 23
Event2011 International Conference on ICT Convergence, ICTC 2011 - Seoul, Korea, Republic of
Duration: 2011 Sep 282011 Sep 30

Other

Other2011 International Conference on ICT Convergence, ICTC 2011
CountryKorea, Republic of
CitySeoul
Period11/9/2811/9/30

Fingerprint

Seed
Experiments

All Science Journal Classification (ASJC) codes

  • Information Systems

Cite this

Kim, Joonsoo ; Choe, Yoonsik ; Kim, Yongoo. / High-quality 2D to 3D video conversion based on robust MRF-based object tracking and reliable graph-cut-based contour refinement. 2011 International Conference on ICT Convergence, ICTC 2011. 2011. pp. 360-365
@inproceedings{b2fc013f75444f4a91a44f178620dc98,
title = "High-quality 2D to 3D video conversion based on robust MRF-based object tracking and reliable graph-cut-based contour refinement",
abstract = "Generation of object-boundary-matched depth map is important to semi-automatic 2D to 3D conversion for high quality perception of converted 3D video. This paper proposes an accurate depth map generation method using an MRF-based initial contour tracking followed by a robust graph-cut-based contour refinement. The proposed MRF-based contour tracking puts a constraint on the variation of angular-radial distance for robust modeling of object motion in the presence of shape deformation. In addition, the proposed contour refinement eliminates ambiguous seeds, which might cause confusion in graph-cut segmentation, resulting in accurate contour in the presence of complex background. Experiment results show the superior tracking performance of the proposed scheme, especially when there is complex background and/or object deformation, indicating the power of the proposed scheme in object-boundary- matched accurate depth map generation.",
author = "Joonsoo Kim and Yoonsik Choe and Yongoo Kim",
year = "2011",
month = "12",
day = "23",
doi = "10.1109/ICTC.2011.6082616",
language = "English",
isbn = "9781457712678",
pages = "360--365",
booktitle = "2011 International Conference on ICT Convergence, ICTC 2011",

}

Kim, J, Choe, Y & Kim, Y 2011, High-quality 2D to 3D video conversion based on robust MRF-based object tracking and reliable graph-cut-based contour refinement. in 2011 International Conference on ICT Convergence, ICTC 2011., 6082616, pp. 360-365, 2011 International Conference on ICT Convergence, ICTC 2011, Seoul, Korea, Republic of, 11/9/28. https://doi.org/10.1109/ICTC.2011.6082616

High-quality 2D to 3D video conversion based on robust MRF-based object tracking and reliable graph-cut-based contour refinement. / Kim, Joonsoo; Choe, Yoonsik; Kim, Yongoo.

2011 International Conference on ICT Convergence, ICTC 2011. 2011. p. 360-365 6082616.

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

TY - GEN

T1 - High-quality 2D to 3D video conversion based on robust MRF-based object tracking and reliable graph-cut-based contour refinement

AU - Kim, Joonsoo

AU - Choe, Yoonsik

AU - Kim, Yongoo

PY - 2011/12/23

Y1 - 2011/12/23

N2 - Generation of object-boundary-matched depth map is important to semi-automatic 2D to 3D conversion for high quality perception of converted 3D video. This paper proposes an accurate depth map generation method using an MRF-based initial contour tracking followed by a robust graph-cut-based contour refinement. The proposed MRF-based contour tracking puts a constraint on the variation of angular-radial distance for robust modeling of object motion in the presence of shape deformation. In addition, the proposed contour refinement eliminates ambiguous seeds, which might cause confusion in graph-cut segmentation, resulting in accurate contour in the presence of complex background. Experiment results show the superior tracking performance of the proposed scheme, especially when there is complex background and/or object deformation, indicating the power of the proposed scheme in object-boundary- matched accurate depth map generation.

AB - Generation of object-boundary-matched depth map is important to semi-automatic 2D to 3D conversion for high quality perception of converted 3D video. This paper proposes an accurate depth map generation method using an MRF-based initial contour tracking followed by a robust graph-cut-based contour refinement. The proposed MRF-based contour tracking puts a constraint on the variation of angular-radial distance for robust modeling of object motion in the presence of shape deformation. In addition, the proposed contour refinement eliminates ambiguous seeds, which might cause confusion in graph-cut segmentation, resulting in accurate contour in the presence of complex background. Experiment results show the superior tracking performance of the proposed scheme, especially when there is complex background and/or object deformation, indicating the power of the proposed scheme in object-boundary- matched accurate depth map generation.

UR - http://www.scopus.com/inward/record.url?scp=83755220058&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=83755220058&partnerID=8YFLogxK

U2 - 10.1109/ICTC.2011.6082616

DO - 10.1109/ICTC.2011.6082616

M3 - Conference contribution

SN - 9781457712678

SP - 360

EP - 365

BT - 2011 International Conference on ICT Convergence, ICTC 2011

ER -