A Rao-Blackwellized Parts-Constellation Tracker
Abstract
We present a method for efficiently tracking objects represented as constellations of parts by integrating out the shape of the model. Parts-based models have been successfully applied to object recognition and tracking. However, the high dimensionality of such models present an obstacle to traditional particle filtering approaches. We can efficiently use parts-based models in a particle filter by applying Rao-Blackwellization to integrate out continuous parameters such as shape. This allows us to maintain multiple hypotheses for the pose of an object without the need to sample in the high-dimensional spaces in which parts-based models live. We present experimental results for a challenging biological tracking task.
Cite
Text
Schindler and Dellaert. "A Rao-Blackwellized Parts-Constellation Tracker." European Conference on Computer Vision, 2006. doi:10.1007/978-3-540-70932-9_14Markdown
[Schindler and Dellaert. "A Rao-Blackwellized Parts-Constellation Tracker." European Conference on Computer Vision, 2006.](https://mlanthology.org/eccv/2006/schindler2006eccv-rao/) doi:10.1007/978-3-540-70932-9_14BibTeX
@inproceedings{schindler2006eccv-rao,
title = {{A Rao-Blackwellized Parts-Constellation Tracker}},
author = {Schindler, Grant and Dellaert, Frank},
booktitle = {European Conference on Computer Vision},
year = {2006},
pages = {178-189},
doi = {10.1007/978-3-540-70932-9_14},
url = {https://mlanthology.org/eccv/2006/schindler2006eccv-rao/}
}