Independent component analysis for understanding multimedia content |
Thomas Kolenda, Lars Kai Hansen, Jan Larsen, Ole Winther
|
| Abstract | This paper focuses on using independent component analysis of combined text and image data from web pages. This has potential for search and retrieval applications in order to retrieve more meaningful and context dependent content. It is demonstrated that using ICA on combined text and image features provides a synergistic effect, i.e., the retrieval classification rates increase if based on multimedia components relative to single media analysis. For this purpose a simple probabilistic supervised classifier which works from unsupervised ICA features is invoked. In addition, we demonstrate the use of the suggested framework for automatic annotation of descriptive key words to images. |
| Keywords | ICA, multimedia signal processing, webmining |
| Type | Conference paper [With referee] |
| Conference | Proceedings of IEEE Workshop on Neural Networks for Signal Processing XII |
| Editors | H. Bourlard, T. Adali, S. Bengio, J. Larsen, and S. Douglas |
| Year | 2002 pp. 757-766 |
| Publisher | IEEE Press |
| Address | Piscataway, New Jersey |
| Note | Data used in publication available in http://www2.compute.dtu.dk/pubdb/p.php?6943 |
| Electronic version(s) | [pdf] |
| BibTeX data | [bibtex] |
| IMM Group(s) | Intelligent Signal Processing |