A real-time visual action-recognition framework for time-varying-geometry objects.
Matthew D. MackayRobert G. FentonBeno BenhabibPublished in: CASE (2010)
Keyphrases
- action recognition
- real time
- video dataset
- computer vision
- human detection
- spatial temporal
- human actions
- static images
- bag of words
- d objects
- action classification
- mid level
- visual information
- depth cameras
- three dimensional
- max margin
- human pose
- recognition of human actions
- video data
- recognizing human actions
- image sequences
- spatio temporal
- bag of features
- low level
- body parts
- spatial relations
- activity recognition