One event, many representations. Mapping action concepts through visual features.
Alessandro PanunziLorenzo GregoriAndrea Amelio RavelliPublished in: LREC (2018)
Keyphrases
- visual features
- human actions
- visual information
- bridge the semantic gap
- image retrieval
- visual content
- image classification
- visual patterns
- keywords
- low level
- low level features
- semantic gap
- image search
- image collections
- visual and textual features
- image annotation
- semantic concepts
- visual data
- event detection
- low level features and high level
- bag of features
- visual appearance
- probabilistic latent semantic analysis
- higher level
- visual properties
- low level visual features
- visual similarity
- key frames
- global features
- video shots
- data extraction
- image content
- motion features
- semantic features
- visual descriptors
- cbir systems
- relevance feedback
- video sequences