Multimodal recognition of visual concepts using histograms of textual concepts and selective weighted late fusion scheme.
Ningning LiuEmmanuel DellandréaLiming ChenChao ZhuYu ZhangCharles-Edmond BichotStéphane BresBruno TellezPublished in: Comput. Vis. Image Underst. (2013)
Keyphrases
- visual concepts
- fusion scheme
- personal identification
- visual content
- biometric systems
- image content
- video content
- learning tasks
- object recognition
- image collections
- object categories
- image annotation
- semantic concepts
- keywords
- visual features
- multimedia
- positive examples
- visual information
- recognition algorithm
- feature extraction
- image representation
- video retrieval
- multi modal
- low level
- iris recognition
- image retrieval
- image matching
- background knowledge
- action recognition