ML Anthology

Factor Graphs for Region-Based Whole-Scene Classification

Matthew R. Boutell, Jiebo Luo, Christopher M. Brown
CVPRW 2006 pp. 104
doi:10.1109/CVPRW.2006.78 /cvprw/2006/boutell2006cvprw-factor/

Abstract

Semantic scene classification is still a challenging problem in computer vision. In contrast to the common approach of using low-level features computed from the scene, our approach uses explicit semantic object detectors and scene configuration models. To overcome faulty semantic detectors, it is critical to develop a region-based, generative model of outdoor scenes based on characteristic objects in the scene and spatial relationships between them. Since a fully connected scene configuration model is intractable, we chose to model pairwise relationships between regions and estimate scene probabilities using loopy belief propagation on a factor graph. We demonstrate the promise of this approach on a set of over 2000 outdoor photographs, comparing it with existing discriminative approaches and those using low-level features.

CVPRW Semantic Scholar

Cite

Text

Boutell et al. "Factor Graphs for Region-Based Whole-Scene Classification." IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2006. doi:10.1109/CVPRW.2006.78

Markdown

[Boutell et al. "Factor Graphs for Region-Based Whole-Scene Classification." IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2006.](https://mlanthology.org/cvprw/2006/boutell2006cvprw-factor/) doi:10.1109/CVPRW.2006.78

BibTeX

@inproceedings{boutell2006cvprw-factor,
  title     = {{Factor Graphs for Region-Based Whole-Scene Classification}},
  author    = {Boutell, Matthew R. and Luo, Jiebo and Brown, Christopher M.},
  booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops},
  year      = {2006},
  pages     = {104},
  doi       = {10.1109/CVPRW.2006.78},
  url       = {https://mlanthology.org/cvprw/2006/boutell2006cvprw-factor/}
}