MDL Patch Correspondences on Unlabeled Images with Occlusions
(2008) IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops (CVPR Workshops), 2008 p.999-1006- Abstract
- Automatic construction of Shape and Appearance Models from examples
via establishing correspondences across the training set has been successful in the last decades.
One successful measure for establishing correspondences of high quality is minimum description length (MDL).
In other approaches it has been shown that parts+geometry models which model the appearance of parts of the object and the geometric relation between the parts
have been successful for automatic model building.
In this paper it is shown how to fuse the above approaches and use MDL to
fully automatically build optimal parts+geometry models from unlabeled images.
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/1219496
- author
- Karlsson, Johan
LU
and Åström, Karl
LU
- organization
- publishing date
- 2008
- type
- Chapter in Book/Report/Conference proceeding
- publication status
- published
- subject
- keywords
- computational geometry, image processing, MDL patch correspondence, unlabeled images, occlusions, automatic construction, shape model, appearance model, training set, minimum description length, automatic model building
- host publication
- 2008 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops (CVPR Workshops)
- pages
- 999 - 1006
- conference name
- IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops (CVPR Workshops), 2008
- conference location
- Anchorage, Alaska, United States
- conference dates
- 2008-06-23 - 2008-06-28
- external identifiers
-
- wos:000260371900136
- scopus:51849093425
- ISBN
- 978-1-4244-2339-2
- DOI
- 10.1109/CVPRW.2008.4563085
- language
- English
- LU publication?
- yes
- id
- 0b858499-e8be-4388-9938-489eb96a9665 (old id 1219496)
- date added to LUP
- 2016-04-04 14:33:21
- date last changed
- 2022-01-30 02:11:50
@inproceedings{0b858499-e8be-4388-9938-489eb96a9665, abstract = {{Automatic construction of Shape and Appearance Models from examples <br/><br> via establishing correspondences across the training set has been successful in the last decades.<br/><br> One successful measure for establishing correspondences of high quality is minimum description length (MDL).<br/><br> In other approaches it has been shown that parts+geometry models which model the appearance of parts of the object and the geometric relation between the parts<br/><br> have been successful for automatic model building.<br/><br> In this paper it is shown how to fuse the above approaches and use MDL to<br/><br> fully automatically build optimal parts+geometry models from unlabeled images.}}, author = {{Karlsson, Johan and Åström, Karl}}, booktitle = {{2008 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops (CVPR Workshops)}}, isbn = {{978-1-4244-2339-2}}, keywords = {{computational geometry; image processing; MDL patch correspondence; unlabeled images; occlusions; automatic construction; shape model; appearance model; training set; minimum description length; automatic model building}}, language = {{eng}}, pages = {{999--1006}}, title = {{MDL Patch Correspondences on Unlabeled Images with Occlusions}}, url = {{http://dx.doi.org/10.1109/CVPRW.2008.4563085}}, doi = {{10.1109/CVPRW.2008.4563085}}, year = {{2008}}, }