File Download
There are no files associated with this item.
Supplementary
-
Citations:
- Scopus: 0
- Appears in Collections:
Conference Paper: Competitive mixture of deformable models for pattern classification
Title | Competitive mixture of deformable models for pattern classification |
---|---|
Authors | |
Keywords | Character recognition Computational geometry Computational methods Computer simulation Feature extraction Iterative methods Mathematical models Optimization |
Issue Date | 1996 |
Publisher | IEEE, Computer Society |
Citation | Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 1996, p. 613-618 How to Cite? |
Abstract | Following 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 Identifier | http://hdl.handle.net/10722/65592 |
ISSN | 2023 SCImago Journal Rankings: 10.331 |
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Cheung, KwokWai | en_HK |
dc.contributor.author | Yeung, DitYan | en_HK |
dc.contributor.author | Chin, Roland T | en_HK |
dc.date.accessioned | 2010-08-31T07:16:23Z | - |
dc.date.available | 2010-08-31T07:16:23Z | - |
dc.date.issued | 1996 | en_HK |
dc.identifier.citation | Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 1996, p. 613-618 | en_HK |
dc.identifier.issn | 1063-6919 | en_HK |
dc.identifier.uri | http://hdl.handle.net/10722/65592 | - |
dc.description.abstract | Following 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.language | eng | en_HK |
dc.publisher | IEEE, Computer Society | en_HK |
dc.relation.ispartof | Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition | en_HK |
dc.subject | Character recognition | en_HK |
dc.subject | Computational geometry | en_HK |
dc.subject | Computational methods | en_HK |
dc.subject | Computer simulation | en_HK |
dc.subject | Feature extraction | en_HK |
dc.subject | Iterative methods | en_HK |
dc.subject | Mathematical models | en_HK |
dc.subject | Optimization | en_HK |
dc.title | Competitive mixture of deformable models for pattern classification | en_HK |
dc.type | Conference_Paper | en_HK |
dc.identifier.email | Chin, Roland T: rchin@hku.hk | en_HK |
dc.identifier.authority | Chin, Roland T=rp01300 | en_HK |
dc.description.nature | link_to_subscribed_fulltext | en_HK |
dc.identifier.scopus | eid_2-s2.0-0029703226 | en_HK |
dc.identifier.spage | 613 | en_HK |
dc.identifier.epage | 618 | en_HK |
dc.publisher.place | United States | en_HK |
dc.identifier.scopusauthorid | Cheung, KwokWai=55413672000 | en_HK |
dc.identifier.scopusauthorid | Yeung, DitYan=7103391392 | en_HK |
dc.identifier.scopusauthorid | Chin, Roland T=7102445426 | en_HK |
dc.identifier.issnl | 1063-6919 | - |