%0 Conference Paper %1 He2004 %A He, Xuming %A Zemel, R.S. %A Carreira-Perpinan, M.A. %B Computer Vision and Pattern Recognition, 2004. CVPR 2004. Proceedings of the 2004 IEEE Computer Society Conference on %D 2004 %K Markov classification, conditional contextual databases, features, field, fields fine-resolution image image-label labeling, mapping, multiscale patterns, processes, random resolution, visual %P II-695--II-702 %R 10.1109/CVPR.2004.1315232 %T Multiscale conditional random fields for image labeling %V 2 %X 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.

@inproceedings{He2004, 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.}, added-at = {2009-09-12T19:19:34.000+0200}, author = {He, Xuming and Zemel, R.S. and Carreira-Perpinan, M.A.}, biburl = {https://www.bibsonomy.org/bibtex/2e0d2e969d40f5a694f1a70d07cea1c1b/mozaher}, booktitle = {Computer Vision and Pattern Recognition, 2004. CVPR 2004. Proceedings of the 2004 IEEE Computer Society Conference on}, doi = {10.1109/CVPR.2004.1315232}, file = {01315232.pdf:He2004.pdf:PDF}, interhash = {b85094c91030b55e1c91f7e47ef9d1f6}, intrahash = {e0d2e969d40f5a694f1a70d07cea1c1b}, issn = {1063-6919}, keywords = {Markov classification, conditional contextual databases, features, field, fields fine-resolution image image-label labeling, mapping, multiscale patterns, processes, random resolution, visual}, month = {27 June-2 July}, owner = {mozaher}, pages = {II-695--II-702}, timestamp = {2009-09-12T19:19:39.000+0200}, title = {Multiscale conditional random fields for image labeling}, volume = 2, year = 2004 }