Multi-Class Image Segmentation Using Conditional Random Fields and Global Classification
Abstract
A key aspect of semantic image segmenta- tion is to integrate local and global features for the prediction of local segment labels. We present an approach to multi-class seg- mentation which combines two methods for this integration: a Conditional Random Field (CRF) which couples to local image features and an image classification method which considers global features. The CRF follows the approach of Reynolds & Murphy (2007) and is based on an unsupervised multi scale pre-segmentation of the image into patches, where patch labels correspond to the ran- dom variables of the CRF. The output of the classifier is used to constraint this CRF. We demonstrate and compare the approach on a standard semantic segmentation data set.
Cite
Text
Plath et al. "Multi-Class Image Segmentation Using Conditional Random Fields and Global Classification." International Conference on Machine Learning, 2009. doi:10.1145/1553374.1553479Markdown
[Plath et al. "Multi-Class Image Segmentation Using Conditional Random Fields and Global Classification." International Conference on Machine Learning, 2009.](https://mlanthology.org/icml/2009/plath2009icml-multi/) doi:10.1145/1553374.1553479BibTeX
@inproceedings{plath2009icml-multi,
title = {{Multi-Class Image Segmentation Using Conditional Random Fields and Global Classification}},
author = {Plath, Nils and Toussaint, Marc and Nakajima, Shinichi},
booktitle = {International Conference on Machine Learning},
year = {2009},
pages = {817-824},
doi = {10.1145/1553374.1553479},
url = {https://mlanthology.org/icml/2009/plath2009icml-multi/}
}