Please use this identifier to cite or link to this item:
https://hdl.handle.net/2440/75166
Citations | ||
Scopus | Web of Science® | Altmetric |
---|---|---|
?
|
?
|
Type: | Journal article |
Title: | Minimum Bayes error features for visual recognition |
Author: | Carneiro, G. Vasconcelos, N. |
Citation: | Image and Vision Computing, 2009; 27(1-2):131-140 |
Publisher: | Elsevier Science BV |
Issue Date: | 2009 |
ISSN: | 0262-8856 1872-8138 |
Statement of Responsibility: | Gustavo Carneiro, Nuno Vasconcelos |
Abstract: | The design of optimal feature sets for visual classification problems is still one of the most challenging topics in the area of computer vision. In this work, we propose a new algorithm that computes optimal features, in the minimum Bayes error sense, for visual recognition tasks. The algorithm now proposed combines the fast convergence rate of feature selection (FS) procedures with the ability of feature extraction (FE) methods to uncover optimal features that are not part of the original basis function set. This leads to solutions that are better than those achievable by either FE or FS alone, in a small number of iterations, making the algorithm scalable in the number of classes of the recognition problem. This property is currently only available for feature extraction methods that are either sub-optimal or optimal under restrictive assumptions that do not hold for generic imagery. Experimental results show significant improvements over these methods, either through much greater robustness to local minima or by achieving significantly faster convergence. © 2006 Elsevier B.V. All rights reserved. |
Keywords: | Visual recognition Feature selection Feature extraction Minimum Bayes error Mixture models Face recognition Texture recognition Object recognition |
Rights: | Copyright © 2006 Elsevier B.V. All rights reserved. |
DOI: | 10.1016/j.imavis.2006.06.008 |
Published version: | http://dx.doi.org/10.1016/j.imavis.2006.06.008 |
Appears in Collections: | Aurora harvest Computer Science publications |
Files in This Item:
There are no files associated with this item.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.