Loading...
Please wait, while we are loading the content...
Similar Documents
Online matrix factorization for multimodal image retrieval.
| Content Provider | CiteSeerX |
|---|---|
| Author | Caicedo, Juan C. González, Fabio A. |
| Abstract | Abstract. In this paper, we propose a method to build an index for image search using multimodal information, that is, using visual features and text data simulta-neously. The method combines both data sources and generates one multimodal representation using latent factor analysis and matrix factorization. One remark-able characteristic of this multimodal representation is that it connects textual and visual content allowing to solve queries with only visual content by implic-itly completing the missing textual content. Another important characteristic of the method is that the multimodal representation is learned online using an ef-ficient stochastic gradient descent formulation. Experiments were conducted in a dataset of 5,000 images to evaluate the convergence speed and search perfor-mance. Experimental results show that the proposed algorithm requires only one pass through the data set to achieve high quality retrieval performance. 1 |
| File Format | |
| Access Restriction | Open |
| Subject Keyword | Online Matrix Factorization Multimodal Image Retrieval Multimodal Representation Visual Content Ef-ficient Stochastic Gradient Descent Formulation Data Source Remark-able Characteristic Latent Factor Analysis Text Data Textual Content Matrix Factorization Image Search Multimodal Information Convergence Speed Search Perfor-mance Visual Feature Important Characteristic Experimental Result High Quality Retrieval Performance |
| Content Type | Text |
| Resource Type | Article |
