Improved action recognition by combining multiple 2D views in the bag-of-words model.
Gertjan J. BurghoutsPieter EendebakHenri BoumaJohan-Martijn ten HovePublished in: AVSS (2013)
Keyphrases
- action recognition
- combining multiple
- human actions
- bag of words
- computer vision
- spatial temporal
- activity recognition
- action classification
- human detection
- body parts
- cluster ensemble
- single view
- recognizing actions
- depth sensors
- recognition of human actions
- static images
- action detection
- bag of features
- recognizing human actions
- video dataset
- view invariant
- object recognition
- action recognition in videos
- action primitives
- human pose
- machine learning
- multiple views
- human activities
- multi class
- independent subspace analysis
- view invariant action recognition