期刊文献+

基于遮挡修补的TV-L1光流算法

TV-L1 optical flow algorithm based on block repair
下载PDF
导出
摘要 当前比较流行的TV-L1光流算法,在不失精确度的前提下,能够利用双向求解机制来降低运算量,但无法有效地处理由间断、遮挡等因素造成的错误光流分量的缺陷。通过前向光流和后向光流的运动一致性理论来判断遮挡区域的光流分量,通过单调递减函数对遮挡区域进行处理,抑制了遮挡区域错误光流对邻域的扩散,提出了同帧邻域光流的横向修补和相邻帧光流的纵向修补。实验表明,该方法能够很好地处理遮挡情况,提高了光流的计算精度。 The TV-L1 optical flow algorithm is relatively popular currently, can reduce the amount of computation with taking advantage of the bidirectional solving mechanism, without losing the accuracy, but cannot effectively process error optical flow component defects caused by interruption, occlusion and other errors. This paper firstly judges optical flow component of occlusion area based on motion coherence of forward and backward optical flow, handles occlusion area by means of Monotone decreasing function, suppresses error light flow of the occlusion area spread to the neighborhood, and presents lateral repairing of the same frame neighborhood optical flow and the longitudinal repairing of the adjacent frames light flow. Experiments show that this method can process occlusion very well, and enhance the calculation precision of optical flow.
出处 《计算机工程与应用》 CSCD 北大核心 2015年第19期164-168,共5页 Computer Engineering and Applications
基金 国家自然科学基金(No.61202439) 湖南省自然科学基金(No.12JJ6057)
关键词 光流 遮挡 光流修补 optical flow occlusion optical flow repairing
  • 相关文献

参考文献15

  • 1Brox T,Bruhn A,Papenberg N,et al.High accuracy optical flow estimation based on a theory for warping[C]//Proceedings of the 8th European Conference on Computer Vision,2004:25-36.
  • 2Papenberg N,Bruhn A,Brox T,et al.Highly accurate optical flow computation with theoretically justified warping[J].International Journal of Computer Vision,2006,67(2):141-158.
  • 3Xiao J,Cheng H,Sawhney H,et al.Bilateral filtering-based optical flow estimation with occlusion detection[C]//Proceedings of the European Conference on Computer Vision,2006,3951:211-224.
  • 4Ince S,Konrad J.Occlusion-aware optical flow estimation[J].IEEE Transactions on Image Processing,2008,17(8):1443-1451.
  • 5Liu C.Beyond pixels:Exploring new representations and applications for motion analysis[D].Cambridge:Massachusettc Institute of Technology,2009.
  • 6Wedel A,Pock T,Zach C,et al.An improved algorithm for TV-L1 optical flow[J].Statistical and Geometrical Approaches to Visual Motion Analysis,2009,12(1):23-45.
  • 7hoi Y J,Lim K W,Ra J B.Improvement on optical flow based video deinterlacing by adopting flow vector and intensity reliabilities[C]//Proceedings of Interrnational Conference on Image Processing,2009:741-744.
  • 8路子赟,唐土生,高隽,沈琳,刘伟.带有局部控制因子的图割光流估计[J].中国图象图形学报,2012,17(7):846-852. 被引量:2
  • 9涂志刚,谢伟,熊淑芬,秦前清.一种高精度的TV-L^1光流算法[J].武汉大学学报(信息科学版),2012,37(4):496-499. 被引量:8
  • 10Shen X H,Wu Y.Exploiting sparsity in dense optical flow[C]//Proceedings of 17th IEEE International Conference on Image Processing,2010:741-744.

二级参考文献4

共引文献7

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部