6 search results - page 1 / 2 » Visual Query Expansion via Incremental Hypernetwork Models o... |
PRICAI
2010
Springer
13 years 2 months ago
2010
Springer
Abstract. Humans can associate vision and language modalities and thus generate mental imagery, i.e. visual images, from linguistic input in an environment of unlimited inflowing i...
PRICAI
2010
Springer
13 years 2 months ago
2010
Springer
Conventional methods for multimodal data retrieval use text-tag based or cross-modal approaches such as tag-image co-occurrence and canonical correlation analysis. Since there are ...
ICCV
2007
IEEE
13 years 10 months ago
2007
IEEE
Given a query image of an object, our objective is to retrieve all instances of that object in a large (1M+) image database. We adopt the bag-of-visual-words architecture which ha...
ICMCS
2006
IEEE
13 years 10 months ago
2006
IEEE
We present methods for improving text search retrieval of visual multimedia content by applying a set of visual models of semantic concepts from a lexicon of concepts deemed relev...
SMI
2008
IEEE
13 years 10 months ago
2008
IEEE
Training a 3D model classifier on a small dataset is very challenging. However, large datasets of partially classified models are now commonly available online. We use an external...