File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Conference Paper: Competitive mixture of deformable models for pattern classification

TitleCompetitive mixture of deformable models for pattern classification
Authors
KeywordsCharacter recognition
Computational geometry
Computational methods
Computer simulation
Feature extraction
Iterative methods
Mathematical models
Optimization
Issue Date1996
PublisherIEEE, Computer Society
Citation
Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 1996, p. 613-618 How to Cite?
AbstractFollowing the success of applying deformable models to feature extraction, a natural next step is to apply such models to pattern classification. Recently, we have cast a deformable model under a Bayesian framework for classification, giving promising results. However, deformable model methods are computationally expensive due to the required iterative optimization process. The problem is even more severe when there are a large number of models (e.g., for character recognition), because each of them has to deform and match with the input data before a final classification can be derived. In this paper, we propose to combine the deformable models into a mixture, in which the individual models compete with each other to survive the matching process during classification. Models that do not compete well are eliminated early, thus allowing substantial savings in computation. This process of competition-elimination has been applied to handwritten digit recognition in which significant speedup can be achieved without sacrificing recognition accuracy.
Persistent Identifierhttp://hdl.handle.net/10722/65592
ISSN

 

DC FieldValueLanguage
dc.contributor.authorCheung, KwokWaien_HK
dc.contributor.authorYeung, DitYanen_HK
dc.contributor.authorChin, Roland Ten_HK
dc.date.accessioned2010-08-31T07:16:23Z-
dc.date.available2010-08-31T07:16:23Z-
dc.date.issued1996en_HK
dc.identifier.citationProceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 1996, p. 613-618en_HK
dc.identifier.issn1063-6919en_HK
dc.identifier.urihttp://hdl.handle.net/10722/65592-
dc.description.abstractFollowing the success of applying deformable models to feature extraction, a natural next step is to apply such models to pattern classification. Recently, we have cast a deformable model under a Bayesian framework for classification, giving promising results. However, deformable model methods are computationally expensive due to the required iterative optimization process. The problem is even more severe when there are a large number of models (e.g., for character recognition), because each of them has to deform and match with the input data before a final classification can be derived. In this paper, we propose to combine the deformable models into a mixture, in which the individual models compete with each other to survive the matching process during classification. Models that do not compete well are eliminated early, thus allowing substantial savings in computation. This process of competition-elimination has been applied to handwritten digit recognition in which significant speedup can be achieved without sacrificing recognition accuracy.en_HK
dc.languageengen_HK
dc.publisherIEEE, Computer Societyen_HK
dc.relation.ispartofProceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognitionen_HK
dc.subjectCharacter recognitionen_HK
dc.subjectComputational geometryen_HK
dc.subjectComputational methodsen_HK
dc.subjectComputer simulationen_HK
dc.subjectFeature extractionen_HK
dc.subjectIterative methodsen_HK
dc.subjectMathematical modelsen_HK
dc.subjectOptimizationen_HK
dc.titleCompetitive mixture of deformable models for pattern classificationen_HK
dc.typeConference_Paperen_HK
dc.identifier.emailChin, Roland T: rchin@hku.hken_HK
dc.identifier.authorityChin, Roland T=rp01300en_HK
dc.description.naturelink_to_subscribed_fulltexten_HK
dc.identifier.scopuseid_2-s2.0-0029703226en_HK
dc.identifier.spage613en_HK
dc.identifier.epage618en_HK
dc.publisher.placeUnited Statesen_HK
dc.identifier.scopusauthoridCheung, KwokWai=55413672000en_HK
dc.identifier.scopusauthoridYeung, DitYan=7103391392en_HK
dc.identifier.scopusauthoridChin, Roland T=7102445426en_HK

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats