Variational optical flow computation assisted by robust point matching

被引:1
|
作者
Shi, Jinlong [1 ]
Bai, Suqin [1 ]
Shu, Xin [1 ]
机构
[1] Jiangsu Univ Sci & Technol, Sch Comp Sci & Engn, Zhenjiang 212003, Peoples R China
来源
OPTIK | 2015年 / 126卷 / 24期
关键词
Motion estimation; Optical flow; Variational; Point matching; TRAJECTORIES; VIDEO;
D O I
10.1016/j.ijleo.2015.09.115
中图分类号
O43 [光学];
学科分类号
070207 ; 0803 ;
摘要
Lots of optical flow methods are based on coarse-to-fine strategy. However, most of coarse-to-fine optical flow techniques face two important problems: namely (1) large displacements of small structures are usually estimated in an incorrect manner and (2) large affine motions of non-rigid objects cannot be computed accurately. In this paper, we present an optical flow computation methodology that integrates robust point matching into a variational optical flow model to deal with the above-mentioned difficulties. Experimental results are demonstrated for a variety of image sequences to validate the effectiveness. (C) 2015 Elsevier GmbH. All rights reserved.
引用
收藏
页码:5092 / 5099
页数:8
相关论文
共 50 条
  • [1] A ROBUST OPTICAL FLOW COMPUTATION
    Lu Zongqing Xie Weixin Pei Jihong (School of Electronic Engineering
    Journal of Electronics(China), 2007, (05) : 635 - 641
  • [2] A robust optical flow computation under variational illumination based on improved census transformation
    Yuan, Jianying
    Wang, Qiong
    Liu, Jiajia
    Li, Bailin
    Jisuanji Fuzhu Sheji Yu Tuxingxue Xuebao/Journal of Computer-Aided Design and Computer Graphics, 2015, 27 (09): : 1725 - 1733
  • [3] Variational optical flow computation in real time
    Bruhn, A
    Weickert, J
    Feddern, C
    Kohlberger, T
    Schnörr, C
    IEEE TRANSACTIONS ON IMAGE PROCESSING, 2005, 14 (05) : 608 - 615
  • [4] Robust multiresolution computation of optical flow
    Ong, EP
    Spann, M
    1996 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING, CONFERENCE PROCEEDINGS, VOLS 1-6, 1996, : 1938 - 1941
  • [5] Robust Point Set Matching Under Variational Bayesian Framework
    Qu, Han-Bing
    Li, Ji-Chao
    Wang, Jia-Qiang
    Xiang, Lin
    Tao, Hai-Jun
    2014 22ND INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION (ICPR), 2014, : 58 - 63
  • [6] Robust tie-point registration of disparate imaging sensors by matching optical flow
    Schuler, JM
    Howard, JG
    Warren, PR
    Scribner, DA
    VISUAL COMMUNICATIONS AND IMAGE PROCESSING 2002, PTS 1 AND 2, 2002, 4671 : 775 - 783
  • [7] Domain decomposition for parallel variational optical flow computation
    Kohlberger, T
    Schnörr, C
    Bruhn, A
    Weickert, J
    PATTERN RECOGNITION, PROCEEDINGS, 2003, 2781 : 196 - 203
  • [8] Domain decomposition for variational optical-flow computation
    Kohlberger, T
    Schnörr, C
    Bruhn, A
    Weickert, J
    IEEE TRANSACTIONS ON IMAGE PROCESSING, 2005, 14 (08) : 1125 - 1137
  • [9] Two Step Variational Method for Subpixel Optical Flow Computation
    Mochizuki, Yoshihiko
    Kameda, Yusuke
    Imiya, Atsushi
    Sakai, Tomoya
    Imaizumi, Takashi
    ADVANCES IN VISUAL COMPUTING, PT 2, PROCEEDINGS, 2009, 5876 : 1109 - +
  • [10] Robust computation of optical flow field with large motion
    Zhang, ZX
    Li, JZ
    Wei, XQ
    2004 7TH INTERNATIONAL CONFERENCE ON SIGNAL PROCESSING PROCEEDINGS, VOLS 1-3, 2004, : 893 - 896