P-CNN: Pose-based CNN Features for Action Recognition.
Guilhem ChéronIvan LaptevCordelia SchmidPublished in: CoRR (2015)
Keyphrases
- action recognition
- human pose
- static images
- bag of features
- activity recognition
- action classification
- human actions
- recognizing actions
- motion features
- computer vision
- deformable part models
- spatial temporal
- mid level
- image features
- action detection
- body parts
- human detection
- human movements
- bag of words
- feature vectors
- low level
- depth cameras
- feature extraction
- space time interest points
- feature space
- machine learning
- recognition of human actions