Reliable optical flow estimation in motion-blurred regions

Yeong Jun Koh, Chul Lee, Jae Young Sim, Chang-Su Kim

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

Abstract

A robust optical flow estimation algorithm for motion-blurred regions is proposed in this work. We first obtain initial optical flow vectors. Then, we detect motion-blurred regions that yield low contrast, low saturation, and inconsistent optical flow vectors. We replace the optical flow vectors at motion-blurred pixels with reliable vectors at nearby unblurred pixels. To this end, we develop an energy minimization framework. Simulation results demonstrate that the proposed algorithm refines optical flow vectors in motion-blurred regions accurately and provides better performance than conventional algorithms.

Original languageEnglish
Title of host publication2013 IEEE International Workshop on Multimedia Signal Processing, MMSP 2013
Pages396-401
Number of pages6
DOIs
Publication statusPublished - 2013 Dec 1
Event2013 IEEE 15th International Workshop on Multimedia Signal Processing, MMSP 2013 - Pula, Sardinia, Italy
Duration: 2013 Sep 302013 Oct 2

Other

Other2013 IEEE 15th International Workshop on Multimedia Signal Processing, MMSP 2013
CountryItaly
CityPula, Sardinia
Period13/9/3013/10/2

Fingerprint

Optical flows
Pixels

Keywords

  • graph cuts optimization
  • motion blur
  • motion estimation
  • Optical flow estimation

ASJC Scopus subject areas

  • Signal Processing

Cite this

Koh, Y. J., Lee, C., Sim, J. Y., & Kim, C-S. (2013). Reliable optical flow estimation in motion-blurred regions. In 2013 IEEE International Workshop on Multimedia Signal Processing, MMSP 2013 (pp. 396-401). [6659321] https://doi.org/10.1109/MMSP.2013.6659321

Reliable optical flow estimation in motion-blurred regions. / Koh, Yeong Jun; Lee, Chul; Sim, Jae Young; Kim, Chang-Su.

2013 IEEE International Workshop on Multimedia Signal Processing, MMSP 2013. 2013. p. 396-401 6659321.

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

Koh, YJ, Lee, C, Sim, JY & Kim, C-S 2013, Reliable optical flow estimation in motion-blurred regions. in 2013 IEEE International Workshop on Multimedia Signal Processing, MMSP 2013., 6659321, pp. 396-401, 2013 IEEE 15th International Workshop on Multimedia Signal Processing, MMSP 2013, Pula, Sardinia, Italy, 13/9/30. https://doi.org/10.1109/MMSP.2013.6659321
Koh YJ, Lee C, Sim JY, Kim C-S. Reliable optical flow estimation in motion-blurred regions. In 2013 IEEE International Workshop on Multimedia Signal Processing, MMSP 2013. 2013. p. 396-401. 6659321 https://doi.org/10.1109/MMSP.2013.6659321
Koh, Yeong Jun ; Lee, Chul ; Sim, Jae Young ; Kim, Chang-Su. / Reliable optical flow estimation in motion-blurred regions. 2013 IEEE International Workshop on Multimedia Signal Processing, MMSP 2013. 2013. pp. 396-401
@inproceedings{0fa8068f42654673ad92dc2f3d9e61e2,
title = "Reliable optical flow estimation in motion-blurred regions",
abstract = "A robust optical flow estimation algorithm for motion-blurred regions is proposed in this work. We first obtain initial optical flow vectors. Then, we detect motion-blurred regions that yield low contrast, low saturation, and inconsistent optical flow vectors. We replace the optical flow vectors at motion-blurred pixels with reliable vectors at nearby unblurred pixels. To this end, we develop an energy minimization framework. Simulation results demonstrate that the proposed algorithm refines optical flow vectors in motion-blurred regions accurately and provides better performance than conventional algorithms.",
keywords = "graph cuts optimization, motion blur, motion estimation, Optical flow estimation",
author = "Koh, {Yeong Jun} and Chul Lee and Sim, {Jae Young} and Chang-Su Kim",
year = "2013",
month = "12",
day = "1",
doi = "10.1109/MMSP.2013.6659321",
language = "English",
isbn = "9781479901258",
pages = "396--401",
booktitle = "2013 IEEE International Workshop on Multimedia Signal Processing, MMSP 2013",

}

TY - GEN

T1 - Reliable optical flow estimation in motion-blurred regions

AU - Koh, Yeong Jun

AU - Lee, Chul

AU - Sim, Jae Young

AU - Kim, Chang-Su

PY - 2013/12/1

Y1 - 2013/12/1

N2 - A robust optical flow estimation algorithm for motion-blurred regions is proposed in this work. We first obtain initial optical flow vectors. Then, we detect motion-blurred regions that yield low contrast, low saturation, and inconsistent optical flow vectors. We replace the optical flow vectors at motion-blurred pixels with reliable vectors at nearby unblurred pixels. To this end, we develop an energy minimization framework. Simulation results demonstrate that the proposed algorithm refines optical flow vectors in motion-blurred regions accurately and provides better performance than conventional algorithms.

AB - A robust optical flow estimation algorithm for motion-blurred regions is proposed in this work. We first obtain initial optical flow vectors. Then, we detect motion-blurred regions that yield low contrast, low saturation, and inconsistent optical flow vectors. We replace the optical flow vectors at motion-blurred pixels with reliable vectors at nearby unblurred pixels. To this end, we develop an energy minimization framework. Simulation results demonstrate that the proposed algorithm refines optical flow vectors in motion-blurred regions accurately and provides better performance than conventional algorithms.

KW - graph cuts optimization

KW - motion blur

KW - motion estimation

KW - Optical flow estimation

UR - http://www.scopus.com/inward/record.url?scp=84892508174&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84892508174&partnerID=8YFLogxK

U2 - 10.1109/MMSP.2013.6659321

DO - 10.1109/MMSP.2013.6659321

M3 - Conference contribution

AN - SCOPUS:84892508174

SN - 9781479901258

SP - 396

EP - 401

BT - 2013 IEEE International Workshop on Multimedia Signal Processing, MMSP 2013

ER -