Investigation of different skeleton features for CNN-based 3D action recognition.
Zewei DingPichao WangPhilip O. OgunbonaWanqing LiPublished in: ICME Workshops (2017)
Keyphrases
- action recognition
- activity recognition
- spatio temporal interest points
- bag of words
- bag of features
- human detection
- static images
- human actions
- spatial temporal
- mid level
- action classification
- space time interest points
- computer vision
- deformable part models
- depth cameras
- feature vectors
- motion features
- recognizing actions
- recognition of human actions
- spatial information
- human movements
- recognizing human actions
- feature extraction
- feature detection
- human pose
- low level
- video dataset
- feature space
- video sequences
- independent subspace analysis
- high level