Incorporating Semantic Attention in Video Description Generation.
Natsuda LaokulratNaoaki OkazakiHideki NakayamaPublished in: LREC (2018)
Keyphrases
- high level
- action descriptions
- semantic concepts
- semantic description
- textual descriptions
- video sequences
- video data
- video event
- video analysis
- semantic knowledge
- real time
- sports video
- video streams
- semantic information
- video content
- video clips
- semantic web
- semantic video
- semantic representation
- natural language
- multimedia
- semantic labels
- semantic video retrieval
- video frames
- semantic annotation
- key frames
- video retrieval
- multi modal
- space time
- content based video retrieval
- content description
- online video
- spatial and temporal
- visual attention
- semantic search
- video annotation
- event recognition
- visual saliency
- semantic similarity
- semantic content
- semantic web technologies
- low level
- symbolic descriptions
- audio visual content
- knowledge base