A Local 3-D Motion Descriptor for Multi-View Human Action Recognition from 4-D Spatio-Temporal Interest Points.
Michael B. HolteBhaskar ChakrabortyJordi GonzàlezThomas B. MoeslundPublished in: IEEE J. Sel. Top. Signal Process. (2012)
Keyphrases
- action recognition
- spatio temporal interest points
- multi view
- human actions
- three dimensional
- single view
- view invariant
- multiple views
- bag of words
- activity recognition
- action classification
- depth map
- d objects
- semi supervised
- computer vision
- visual hull
- human pose
- human activities
- image sequences
- human motion
- body parts
- structure from motion
- motion capture
- video sequences
- multiple viewpoints
- feature points
- pairwise