Action recognition by orthogonalized subspaces of local spatio-temporal features.
Bisser RaytchevRyosuke ShigenakaToru TamakiKazufumi KanedaPublished in: ICIP (2013)
Keyphrases
- action recognition
- human actions
- activity recognition
- bag of words
- spatial temporal
- computer vision
- body parts
- human detection
- action classification
- static images
- human pose
- recognizing actions
- mid level
- recognizing human actions
- video dataset
- depth sensors
- view invariant
- human activities
- recognition of human actions
- bag of features
- independent subspace analysis
- depth cameras
- max margin
- multi view
- object detection
- spatio temporal