A hybrid architecture based on CNN for cross-modal semantic instance annotation.
Yongzhe ZhengZhixin LiCanlong ZhangPublished in: Multim. Tools Appl. (2018)
Keyphrases
- hybrid architecture
- cross modal
- multi modal
- semantic annotation
- image retrieval
- visual similarity
- image annotation
- semantic concepts
- multimedia retrieval
- automatic image annotation
- visual recognition
- semantic web
- semantic information
- high level
- agent architecture
- multimedia databases
- low level features
- visual concepts
- visual data
- semantic similarity
- active learning
- text retrieval
- process management
- information systems
- information retrieval
- image database
- low level
- multimedia information retrieval
- multi agent
- multimedia
- metadata