Hierarchical Multimodal LSTM for Dense Visual-Semantic Embedding.
Zhenxing NiuMo ZhouLe WangXinbo GaoGang HuaPublished in: ICCV (2017)
Keyphrases
- semantic content
- semantic information
- visual information
- multi modal
- visual features
- cross modal
- multimodal information
- semantically relevant
- semantic annotation
- recurrent neural networks
- object based visual attention
- high level
- semantic context
- semantic network
- semantic similarity
- semantic web
- domain specific
- natural language
- hierarchical structure
- semantically related
- semantic description
- semantic space
- coarse to fine
- low level features
- semantic features
- concept hierarchy
- domain ontology
- single modality
- gaussian process latent variable models
- low level