Name
Abstract | ||
|---|---|---|
Recently, generic object recognition (automatic image annotation) that achieves human-like vision using a computer has being looked to for use in robot vision, automatic categorization of images, and retrieval of images. For the annotation, semi-supervised learning, which incorporates a large amount of unsupervised training data (unlabeled data) along with a small amount of supervised data (labeled data), is expected to be an effective tool as it reduces the burden of manual annotation. However, some unlabeled data in semi-supervised models contains outliers that negatively affect the parameter estimation on the training stage. Such outliers often cause the over-fitting problem especially when a small amount of training data is used. In this paper, we propose a practical method to prevent the over-fitting in semi-supervised learning, suppressing existing outliers by sparse representation. In our experiments we got 4 points improvement comparing conventional semi-supervised methods, SemiNB and TSVM. |
Year | DOI | Venue |
|---|---|---|
2013 | 10.1109/ICASSP.2013.6638020 | Acoustics, Speech and Signal Processing |
Keywords | Field | DocType |
learning (artificial intelligence),object recognition,generic object recognition,image automatic categorization,image retrieval,labeled data,outliers suppression,parameter estimation,robot vision,semisupervised image annotation,sparse representation,supervised data,unlabeled data,unsupervised training data,Object recognition,automatic anotation,semi-supervised learning,sparse representation | Categorization,Automatic image annotation,Semi-supervised learning,Annotation,Pattern recognition,Computer science,Sparse approximation,Image retrieval,Unsupervised learning,Artificial intelligence,Machine learning,Cognitive neuroscience of visual object recognition | Conference |
ISSN | Citations | PageRank |
1520-6149 | 0 | 0.34 |
References | Authors | |
4 | 4 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Toru Nakashika | 1 | 81 | 13.60 |
| Takeshi Okumura | 2 | 0 | 0.68 |
| Tetsuya Takiguchi | 3 | 85 | 8.77 |
| Yasuo Ariki | 4 | 519 | 88.94 |