Automatic feature point correspondences and shape analysis with missing data and outliers using MDL
(2007) 15th Scandinavian Image Analysis Conference 4522. p.21-30- Abstract
- Automatic construction of shape models from examples has recently been the focus of intense research. These methods have proved to be useful for shape segmentation, tracking, recognition and shape understanding. In this paper we discuss automatic landmark selection and correspondence determination from a discrete set of landmarks, typically obtained by feature extraction. The set of landmarks may include both outliers and missing data. Our framework has a solid theoretical basis using principles of minimal description length (MDL). In order to exploit these ideas, new non-heuristic methods for (i) principal component analysis and (ii) procrustes mean are derived - as a consequence of the modelling principle. The resulting MDL criterion is... (More)
- Automatic construction of shape models from examples has recently been the focus of intense research. These methods have proved to be useful for shape segmentation, tracking, recognition and shape understanding. In this paper we discuss automatic landmark selection and correspondence determination from a discrete set of landmarks, typically obtained by feature extraction. The set of landmarks may include both outliers and missing data. Our framework has a solid theoretical basis using principles of minimal description length (MDL). In order to exploit these ideas, new non-heuristic methods for (i) principal component analysis and (ii) procrustes mean are derived - as a consequence of the modelling principle. The resulting MDL criterion is optimised over both discrete and continuous decision variables. The algorithms have been implemented and tested on the problem of automatic shape extraction from feature points in image sequences. (Less)
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/643262
- author
- Åström, Karl
LU
; Karlsson, Johan LU ; Enqvist, Olof LU ; Ericsson, Anders LU and Kahl, Fredrik LU
- organization
- publishing date
- 2007
- type
- Chapter in Book/Report/Conference proceeding
- publication status
- published
- subject
- keywords
- minimal description length, automatic construction, image segmentation, image recognition, tracking, feature extraction, image sequence, shape analysis, principal component analysis
- host publication
- Proceedings 15th Scandinavian Image Analysis Conference
- editor
- Ersböll, Bjarne Kjaer and Pedersen, Kim Stenstrup
- volume
- 4522
- pages
- 21 - 30
- publisher
- Springer
- conference name
- 15th Scandinavian Image Analysis Conference
- conference location
- Aalborg, Denmark
- conference dates
- 2007-06-10 - 2007-06-14
- external identifiers
-
- wos:000247364000003
- scopus:38049077020
- ISBN
- ISBN 978-3-540-73039-2
- DOI
- 10.1007/978-3-540-73040-8_3
- language
- English
- LU publication?
- yes
- id
- af5a64c5-c4c9-4e6d-8245-3fc7cb0dd7c3 (old id 643262)
- date added to LUP
- 2016-04-04 10:04:52
- date last changed
- 2022-01-29 19:42:56
@inproceedings{af5a64c5-c4c9-4e6d-8245-3fc7cb0dd7c3, abstract = {{Automatic construction of shape models from examples has recently been the focus of intense research. These methods have proved to be useful for shape segmentation, tracking, recognition and shape understanding. In this paper we discuss automatic landmark selection and correspondence determination from a discrete set of landmarks, typically obtained by feature extraction. The set of landmarks may include both outliers and missing data. Our framework has a solid theoretical basis using principles of minimal description length (MDL). In order to exploit these ideas, new non-heuristic methods for (i) principal component analysis and (ii) procrustes mean are derived - as a consequence of the modelling principle. The resulting MDL criterion is optimised over both discrete and continuous decision variables. The algorithms have been implemented and tested on the problem of automatic shape extraction from feature points in image sequences.}}, author = {{Åström, Karl and Karlsson, Johan and Enqvist, Olof and Ericsson, Anders and Kahl, Fredrik}}, booktitle = {{Proceedings 15th Scandinavian Image Analysis Conference}}, editor = {{Ersböll, Bjarne Kjaer and Pedersen, Kim Stenstrup}}, isbn = {{ISBN 978-3-540-73039-2}}, keywords = {{minimal description length; automatic construction; image segmentation; image recognition; tracking; feature extraction; image sequence; shape analysis; principal component analysis}}, language = {{eng}}, pages = {{21--30}}, publisher = {{Springer}}, title = {{Automatic feature point correspondences and shape analysis with missing data and outliers using MDL}}, url = {{http://dx.doi.org/10.1007/978-3-540-73040-8_3}}, doi = {{10.1007/978-3-540-73040-8_3}}, volume = {{4522}}, year = {{2007}}, }