Nonlinear matrix diffusion for optic flow estimation

Thomas Brox, J. Weickert
Pattern Recognition (Proc. DAGM), Springer, LNCS, Vol.2449: 446-453, Sept. 2002
Abstract: In this paper we present a method for nonlinear diffusion of matrix-valued data. We adapt this technique to the well-known linear structure tensor in order to develop a new nonlinear structure tensor. It is then used to improve the optic flow estimation methods of Lucas and Kanade and its spatio-temporal variant of Bigün et al.. Our experiments show that the nonlinear structure tensor leads to a better preservation of discontinuities in the optic flow field.

Other associated files : brox_dagm02.pdf [1.3MB]  

Images and movies


BibTex reference

  author       = "T. Brox and J. Weickert",
  title        = "Nonlinear matrix diffusion for optic flow estimation",
  booktitle    = "Pattern Recognition (Proc. DAGM)",
  series       = "Lecture Notes in Computer Science",
  volume       = "2449",
  pages        = "446-453",
  month        = "Sept.",
  year         = "2002",
  publisher    = "Springer",
  url          = "http://lmb.informatik.uni-freiburg.de/Publications/2002/Bro02"

Other publications in the database