Enregistré dans:
| Auteur principal: | |
|---|---|
| Format: | Artículo científico |
| Langue: | en |
| Publié: |
Instituto Politécnico Nacional
2003
|
| Sujets: | |
| Accès en ligne: | https://www.redalyc.org/articulo.oa?id=61570206 |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
Table des matières:
- Face Recognition Using Unlabeled Data Carmen Martínez Olac Fuentes Computación Face recognition systems can normally attain good accuracy when they are provided with a large set oftraining examples. However, when a large training set is not available, their performance is commonly poor.In this work we describe a method for face recognition that achieves good results when only a very smalltraining set is available (one image per person). The method is based on augmenting the original training setwith previously unlabeled data (that is, face images for which the identity of the person is not known).Initially, we apply the well-known eigenfaces technique to reduce the dimensionality of the image space,then we perform an iterative process, classifying all the unlabeled data with an ensemble of classifiers builtfrom the current training set, and appending to the training set the previously unlabeled examples that arebelieved to be correctly classified with a high confidence level, according to the ensemble.We experimented with ensembles based on the k-nearest neighbors, feed forward artificial neural networksand locally weighted linear regression learning algorithms. Our experimental results show that usingunlabeled data improves the accuracy in all cases. The best accuracy, 92.07%, was obtained with locallyweighted linear regression using 30 eigenfaces and appending 3 examples of every class in each iteration. Incontrast, using only labeled data, an accuracy of only 34.81% was obtained. 2003 artículo científico 1405-5546 https://www.redalyc.org/articulo.oa?id=61570206 en http://www.redalyc.org/revista.oa?id=615 Computación y Sistemas application/pdf Instituto Politécnico Nacional Computación y Sistemas (México) Num.2 Vol.7