Loading...
Please wait, while we are loading the content...
Similar Documents
Topic Regression Multi-Modal Latent Dirichlet Allocation for Image Annotation
| Content Provider | CiteSeerX |
|---|---|
| Author | Putthividhya, Duangmanee Attias, Hagai T. Nagarajan, Srikantan S. |
| Abstract | We present topic-regression multi-modal Latent Dirich-let Allocation (tr-mmLDA), a novel statistical topic model for the task of image and video annotation. At the heart of our new annotation model lies a novel latent variable re-gression approach to capture correlations between image or video features and annotation texts. Instead of sharing a set of latent topics between the 2 data modalities as in the formulation of correspondence LDA in [2], our approach introduces a regression module to correlate the 2 sets of topics, which captures more general forms of association and allows the number of topics in the 2 data modalities to be different. We demonstrate the power of tr-mmLDA on 2 standard annotation datasets: a 5000-image subset of COREL and a 2687-image LabelMe dataset. The proposed association model shows improved performance over cor-respondence LDA as measured by caption perplexity. 1. |
| File Format | |
| Access Restriction | Open |
| Subject Keyword | Regression Multi-modal Latent Dirichlet Allocation Image Annotation Data Modality 2687-image Labelme Dataset Caption Perplexity Cor-respondence Lda Correspondence Lda Video Feature Regression Module Novel Latent Variable Re-gression Approach Association Model Novel Statistical Topic Model Standard Annotation Datasets General Form Annotation Text Latent Topic Video Annotation 5000-image Subset New Annotation Model |
| Content Type | Text |
