Human action recognition in crowded surveillance video sequences by using features taken from key-point trajectories.
Masaki TakahashiMasahide NaemuraMahito FujiiShin'ichi SatohPublished in: CVPR Workshops (2011)
Keyphrases
- action recognition
- human actions
- space time interest points
- video sequences
- body parts
- motion capture data
- motion history images
- human detection
- bag of words
- human activities
- activity recognition
- low level
- human object interactions
- computer vision
- feature extraction
- video shots
- human pose
- image features
- video database
- spatial information
- human motion
- spatio temporal
- viewpoint
- point trajectories
- image sequences
- three dimensional