MDL Patch Correspondences on Unlabeled Images with Occlusions
Karlsson, Johan; Åström, Karl (2008). MDL Patch Correspondences on Unlabeled Images with Occlusions 2008 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops (CVPR Workshops), 999 - 1006. IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops (CVPR Workshops), 2008. Anchorage, Alaska, United States
Conference Proceeding/Paper
|
Published
|
English
Authors:
Karlsson, Johan
;
Åström, Karl
Department:
Mathematics (Faculty of Engineering)
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.
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
Cite this