A Bottom-up and Top-Down Optimization Framework for Learning a Compositional Hierarchy of Object Classes
Abstract
Learning hierarchical representations of object structure in a bottom-up manner faces several difficult issues. First, we are dealing with a very large number of potential feature aggregations. Furthermore, the set of features the algorithm learns at each layer directly influences the expressiveness of the compositional layers that work on top of them. However, we cannot ensure the usefulness of a particular local feature for object class representation based solely on the local statistics. This can only be done when more global, object-wise information is taken into account.
Cite
Text
Fidler et al. "A Bottom-up and Top-Down Optimization Framework for Learning a Compositional Hierarchy of Object Classes." IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2009. doi:10.1109/CVPRW.2009.5204327Markdown
[Fidler et al. "A Bottom-up and Top-Down Optimization Framework for Learning a Compositional Hierarchy of Object Classes." IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2009.](https://mlanthology.org/cvprw/2009/fidler2009cvprw-bottomup/) doi:10.1109/CVPRW.2009.5204327BibTeX
@inproceedings{fidler2009cvprw-bottomup,
title = {{A Bottom-up and Top-Down Optimization Framework for Learning a Compositional Hierarchy of Object Classes}},
author = {Fidler, Sanja and Boben, Marko and Leonardis, Ales},
booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops},
year = {2009},
pages = {3},
doi = {10.1109/CVPRW.2009.5204327},
url = {https://mlanthology.org/cvprw/2009/fidler2009cvprw-bottomup/}
}