A region-based selective optical flow back-projection for genuine motion vector estimation

Pau-Choo Chung, C. L. Huang, E. L. Chen

Research output: Contribution to journalArticle

9 Citations (Scopus)

Abstract

Motion vector plays one significant feature in moving object segmentation. However, the motion vector in this application is required to represent the actual motion displacement, rather than regions of visually significant similarity. In this paper, region-based selective optical flow back-projection (RSOFB) which back-projects optical flows in a region to restore the region's motion vector from gradient-based optical flows, is proposed to obtain genuine motion displacement. The back-projection is performed based on minimizing the projection mean square errors of the motion vector on gradient directions. As optical flows of various magnitudes and directions provide various degrees of reliability in the genuine motion restoration, the optical flows to be used in the RSOFB are optimally selected based on their sensitivity to noises and their tendency in causing motion estimation errors. In this paper a deterministic solution is also derived for performing the minimization and obtaining the genuine motion magnitude and motion direction.

Original languageEnglish
Pages (from-to)1066-1077
Number of pages12
JournalPattern Recognition
Volume40
Issue number3
DOIs
Publication statusPublished - 2007 Mar 1

Fingerprint

Optical flows
Motion estimation
Mean square error
Restoration

All Science Journal Classification (ASJC) codes

  • Computer Vision and Pattern Recognition
  • Signal Processing
  • Electrical and Electronic Engineering

Cite this

@article{bbfb9d68822b4cf39efd3832ecf5fb9f,
title = "A region-based selective optical flow back-projection for genuine motion vector estimation",
abstract = "Motion vector plays one significant feature in moving object segmentation. However, the motion vector in this application is required to represent the actual motion displacement, rather than regions of visually significant similarity. In this paper, region-based selective optical flow back-projection (RSOFB) which back-projects optical flows in a region to restore the region's motion vector from gradient-based optical flows, is proposed to obtain genuine motion displacement. The back-projection is performed based on minimizing the projection mean square errors of the motion vector on gradient directions. As optical flows of various magnitudes and directions provide various degrees of reliability in the genuine motion restoration, the optical flows to be used in the RSOFB are optimally selected based on their sensitivity to noises and their tendency in causing motion estimation errors. In this paper a deterministic solution is also derived for performing the minimization and obtaining the genuine motion magnitude and motion direction.",
author = "Pau-Choo Chung and Huang, {C. L.} and Chen, {E. L.}",
year = "2007",
month = "3",
day = "1",
doi = "10.1016/j.patcog.2006.06.019",
language = "English",
volume = "40",
pages = "1066--1077",
journal = "Pattern Recognition",
issn = "0031-3203",
publisher = "Elsevier Limited",
number = "3",

}

A region-based selective optical flow back-projection for genuine motion vector estimation. / Chung, Pau-Choo; Huang, C. L.; Chen, E. L.

In: Pattern Recognition, Vol. 40, No. 3, 01.03.2007, p. 1066-1077.

Research output: Contribution to journalArticle

TY - JOUR

T1 - A region-based selective optical flow back-projection for genuine motion vector estimation

AU - Chung, Pau-Choo

AU - Huang, C. L.

AU - Chen, E. L.

PY - 2007/3/1

Y1 - 2007/3/1

N2 - Motion vector plays one significant feature in moving object segmentation. However, the motion vector in this application is required to represent the actual motion displacement, rather than regions of visually significant similarity. In this paper, region-based selective optical flow back-projection (RSOFB) which back-projects optical flows in a region to restore the region's motion vector from gradient-based optical flows, is proposed to obtain genuine motion displacement. The back-projection is performed based on minimizing the projection mean square errors of the motion vector on gradient directions. As optical flows of various magnitudes and directions provide various degrees of reliability in the genuine motion restoration, the optical flows to be used in the RSOFB are optimally selected based on their sensitivity to noises and their tendency in causing motion estimation errors. In this paper a deterministic solution is also derived for performing the minimization and obtaining the genuine motion magnitude and motion direction.

AB - Motion vector plays one significant feature in moving object segmentation. However, the motion vector in this application is required to represent the actual motion displacement, rather than regions of visually significant similarity. In this paper, region-based selective optical flow back-projection (RSOFB) which back-projects optical flows in a region to restore the region's motion vector from gradient-based optical flows, is proposed to obtain genuine motion displacement. The back-projection is performed based on minimizing the projection mean square errors of the motion vector on gradient directions. As optical flows of various magnitudes and directions provide various degrees of reliability in the genuine motion restoration, the optical flows to be used in the RSOFB are optimally selected based on their sensitivity to noises and their tendency in causing motion estimation errors. In this paper a deterministic solution is also derived for performing the minimization and obtaining the genuine motion magnitude and motion direction.

UR - http://www.scopus.com/inward/record.url?scp=33750495145&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33750495145&partnerID=8YFLogxK

U2 - 10.1016/j.patcog.2006.06.019

DO - 10.1016/j.patcog.2006.06.019

M3 - Article

VL - 40

SP - 1066

EP - 1077

JO - Pattern Recognition

JF - Pattern Recognition

SN - 0031-3203

IS - 3

ER -