Exploring the multimodal information from video content using deep learning features of appearance, audio and action for video recommendation.
Adolfo AlmeidaJohan Pieter de VilliersAllan De FreitasMergandran VelayudanPublished in: CoRR (2020)
Keyphrases
- video content
- multimodal information
- video data
- deep learning
- video clips
- video streams
- key frames
- video sequences
- visual data
- video frames
- video retrieval
- multimedia
- video analysis
- video material
- video shots
- feature vectors
- audio features
- unsupervised learning
- low level
- multimedia content
- three dimensional
- feature extraction
- machine learning
- action recognition
- co occurrence