Multiscale Conditional Random Fields for Image Labeling
Abstract
We propose an approach to include contextual features for labeling images, in which each pixel is assigned to one of a finite set of labels. The features are incorporated into a probabilistic framework, which combines the outputs of several components. Components differ in the information they encode. Some focus on the image-label mapping, while others focus solely on patterns within the label field. Components also differ in their scale, as some focus on fine-resolution patterns while others on coarser, more global structure. A supervised version of the contrastive divergence algorithm is applied to learn these features from labeled image data. We demonstrate performance on two real-world image databases and compare it to a classifier and a Markov random field.
Cite
Text
He et al. "Multiscale Conditional Random Fields for Image Labeling." IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2004. doi:10.1109/CVPR.2004.173Markdown
[He et al. "Multiscale Conditional Random Fields for Image Labeling." IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2004.](https://mlanthology.org/cvpr/2004/he2004cvpr-multiscale/) doi:10.1109/CVPR.2004.173BibTeX
@inproceedings{he2004cvpr-multiscale,
title = {{Multiscale Conditional Random Fields for Image Labeling}},
author = {He, Xuming and Zemel, Richard S. and Carreira-Perpiñán, Miguel Á.},
booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition},
year = {2004},
pages = {695-702},
doi = {10.1109/CVPR.2004.173},
url = {https://mlanthology.org/cvpr/2004/he2004cvpr-multiscale/}
}