Effective Codebooks for Human Action Representation and Classification in Unconstrained Videos.
Lamberto BallanMarco BertiniAlberto Del BimboLorenzo SeidenariGiuseppe SerraPublished in: IEEE Trans. Multim. (2012)
Keyphrases
- human actions
- action recognition
- recognition of human actions
- view invariant
- action classification
- recognizing human actions
- spatio temporal
- action recognition in videos
- human activities
- video sequences
- spatio temporal interest points
- image representation
- space time
- video database
- bag of words
- human motion
- image classification
- recognizing actions
- feature space
- single view
- decision trees
- multi view
- machine learning
- motion recognition
- activity recognition
- human pose
- feature extraction
- motion features
- motion primitives
- visual features
- action detection
- text classification
- wireless sensor networks
- feature vectors
- computer vision