Header logo is ps

Probabilistic detection and tracking of motion boundaries




We propose a Bayesian framework for representing and recognizing local image motion in terms of two basic models: translational motion and motion boundaries. Motion boundaries are represented using a non-linear generative model that explicitly encodes the orientation of the boundary, the velocities on either side, the motion of the occluding edge over time, and the appearance/disappearance of pixels at the boundary. We represent the posterior probability distribution over the model parameters given the image data using discrete samples. This distribution is propagated over time using a particle filtering algorithm. To efficiently represent such a high-dimensional space we initialize samples using the responses of a low-level motion discontinuity detector. The formulation and computational model provide a general probabilistic framework for motion estimation with multiple, non-linear, models.

Author(s): Black, M. J. and Fleet, D. J.
Journal: Int. J. of Computer Vision
Volume: 38
Number (issue): 3
Pages: 231-245
Year: 2000
Month: July

Department(s): Perceiving Systems
Bibtex Type: Article (article)
Paper Type: Journal

Links: pdf
pdf from publisher


  title = {Probabilistic detection and tracking of motion boundaries},
  author = {Black, M. J. and Fleet, D. J.},
  journal = {Int. J. of Computer Vision},
  volume = {38},
  number = {3},
  pages = {231-245},
  month = jul,
  year = {2000},
  month_numeric = {7}