Title MDL Patch Correspondences on Unlabeled Images
Authors Johan Karlsson, Karl Åström
Publication 19th International Conference on Pattern Recognition, vols 1-6
Year 2008
Pages 3249 - 3253
Document type Conference paper
Conference name International Conference on Pattern Recognition
Conference Date Dec 08-11, 2008
Conference Location Tampa, FL, United States
Status Published
Quality controlled Yes
Language eng
Publisher IEEE Press
Abstract English Automatic construction of Shape and Appearance Models from examples <br> via establishing correspondences across the training set has been successful in the last decades.<br> One successful measure for establishing correspondences of high quality is minimum description length (MDL).<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> have been successful for automatic model building.<br> In this paper it is shown how to fuse the above approaches and use MDL to<br> fully automatically build optimal parts+geometry models from unlabeled <br> images.
ISBN/ISSN/Other ISSN: 1051-4651
ISBN: 978-1-4244-2174-9

Questions: webmaster
Last update: 2013-04-11

Centre for Mathematical Sciences, Box 118, SE-22100, Lund. Telefon: +46 46-222 00 00 (vx)