ML Anthology

A Comparison of Stochastic and Deterministic Solution Methods in Bayesian Estimation of 2-D Motion

Janusz Konrad, Eric Dubois
ECCV 1990 pp. 149-160
doi:10.1007/BFB0014861 /eccv/1990/konrad1990eccv-comparison/

Abstract

A new stochastic motion estimation method based on the Maximum A Posteriori Probability (MAP) criterion is developed. Deterministic algorithms approximating the MAP estimation over discrete and continuous state spaces are proposed. These approximations result in known motion estimation algorithms. The theoretical superiority of the stochastic algorithms over deterministic approximations in locating the global optimum is confirmed experimentally.

PDF ECCV Semantic Scholar

Cite

Text

Konrad and Dubois. "A Comparison of Stochastic and Deterministic Solution Methods in Bayesian Estimation of 2-D Motion." European Conference on Computer Vision, 1990. doi:10.1007/BFB0014861

Markdown

[Konrad and Dubois. "A Comparison of Stochastic and Deterministic Solution Methods in Bayesian Estimation of 2-D Motion." European Conference on Computer Vision, 1990.](https://mlanthology.org/eccv/1990/konrad1990eccv-comparison/) doi:10.1007/BFB0014861

BibTeX

@inproceedings{konrad1990eccv-comparison,
  title     = {{A Comparison of Stochastic and Deterministic Solution Methods in Bayesian Estimation of 2-D Motion}},
  author    = {Konrad, Janusz and Dubois, Eric},
  booktitle = {European Conference on Computer Vision},
  year      = {1990},
  pages     = {149-160},
  doi       = {10.1007/BFB0014861},
  url       = {https://mlanthology.org/eccv/1990/konrad1990eccv-comparison/}
}