Improving acoustic event detection using generalizable visual features and multi-modality modeling.
Po-Sen HuangXiaodan ZhuangMark Hasegawa-JohnsonPublished in: ICASSP (2011)
Keyphrases
- visual features
- event detection
- multi modality
- multi modal
- image classification
- acoustic features
- visual information
- image search
- low level
- low level features
- information theoretic
- visual content
- image annotation
- event recognition
- medical images
- video analysis
- image retrieval
- image registration
- keywords
- semantic concepts
- mutual information
- image collections
- sports video
- activity recognition
- human actions
- multimedia
- video shots
- key frames
- imaging modalities
- semantic content
- web images
- data mining
- input image
- machine learning