Detecting violent content in Hollywood movies by mid-level audio representations.
Esra AcarFrank HopfgartnerSahin AlbayrakPublished in: CBMI (2013)
Keyphrases
- mid level
- action recognition
- visual information
- multimedia
- low level
- visual concepts
- high level
- visual cues
- levels of abstraction
- low level features
- perceptual grouping
- event detection
- visual data
- bag of words
- computer vision
- semantic information
- video data
- activity recognition
- multimedia content
- higher level
- metadata
- visual features
- superpixels
- learning experience
- human activities
- object recognition
- video sequences
- e learning
- information retrieval