Visual and language semantic hybrid enhancement and complementary for video description.
Pengjie TangYunlan TanWenlang LuoPublished in: Neural Comput. Appl. (2022)
Keyphrases
- content based video retrieval
- high level
- semantic concepts
- natural language
- visual features
- visual concepts
- visual data
- video data
- action descriptions
- programming language
- semantic description
- textual descriptions
- semantic labels
- visual cues
- visual information
- video streams
- video content
- semantically equivalent
- video retrieval
- video event
- concept detectors
- video database
- semantic video
- low level
- semantic content
- multimedia data
- semantic representation
- multimedia
- image processing
- context dependent
- video sequences
- conceptual graphs
- sports video
- semantic search
- semantically relevant
- linguistic analysis
- video frames
- semantic context
- semantic network
- semantic representations
- semantic structure
- video annotation
- key frames
- language learning
- video search
- semantic knowledge
- real time
- video clips
- semantic information
- video analysis
- visual analysis
- action language
- low level features
- content description
- space time
- semantic space
- news video
- semantic similarity
- visual input
- visual saliency
- visual query language
- web images
- video shots
- image content
- image classification