Human action recognition based on two-view optical flow in the transformed domain.
Mohamed A. AbdelwahabMoataz M. AbdelwahabPublished in: MWSCAS (2014)
Keyphrases
- action recognition
- optical flow
- static images
- computer vision
- human movements
- human activities
- bag of words
- human actions
- activity recognition
- motion features
- human detection
- motion capture data
- motion history images
- spatial temporal
- recognition of human actions
- body parts
- recognizing human actions
- human object interactions
- human pose
- mid level
- action classification
- depth cameras
- flow field
- low level
- depth sensors
- view invariant
- motion model
- scene flow
- video dataset
- motion estimation
- moving objects
- d scene
- image sequences
- bag of features
- motion field
- machine learning