Filtering Using a Tree-Based Estimator

Abstract

Within this paper a new framework for Bayesian tracking is presented, which approximates the posterior distribution at multiple resolutions. We propose a tree-based representation of the distribution, where the leaves define a partition of the state space with piecewise constant density. The advantage of this representation is that regions with low probability mass can be rapidly discarded in a hierarchical search, and the distribution can be approximated to arbitrary precision. We demonstrate the effectiveness of the technique by using it for tracking 3D articulated and non-rigid motion in front of cluttered background. More specifically, we are interested in estimating the joint angles, position and orientation of a 3D hand model in order to drive an avatar. 1.

Cite

Text

Stenger et al. "Filtering Using a Tree-Based Estimator." IEEE/CVF International Conference on Computer Vision, 2003. doi:10.1109/ICCV.2003.1238467

Markdown

[Stenger et al. "Filtering Using a Tree-Based Estimator." IEEE/CVF International Conference on Computer Vision, 2003.](https://mlanthology.org/iccv/2003/stenger2003iccv-filtering/) doi:10.1109/ICCV.2003.1238467

BibTeX

@inproceedings{stenger2003iccv-filtering,
  title     = {{Filtering Using a Tree-Based Estimator}},
  author    = {Stenger, Bjoern and Thayananthan, Arasanathan and Torr, Philip H. S. and Cipolla, Roberto},
  booktitle = {IEEE/CVF International Conference on Computer Vision},
  year      = {2003},
  pages     = {1063-1070},
  doi       = {10.1109/ICCV.2003.1238467},
  url       = {https://mlanthology.org/iccv/2003/stenger2003iccv-filtering/}
}