On the improvement of human action recognition from depth map sequences using Space-Time Occupancy Patterns.
Antônio Wilson VieiraErickson R. NascimentoGabriel L. OliveiraZicheng LiuMario Fernando Montenegro CamposPublished in: Pattern Recognit. Lett. (2014)
Keyphrases
- space time
- depth map
- action recognition
- human actions
- motion history images
- multi view
- depth information
- d scene
- video sequences
- human activities
- spatio temporal
- high quality
- activity recognition
- stereo matching
- three dimensional
- depth estimation
- motion patterns
- image sequences
- dynamic scenes
- input image
- temporal patterns
- view synthesis
- computer vision
- human motion
- stereo pair
- image processing
- depth data
- time of flight
- stereo camera
- body parts
- high resolution
- super resolution
- stereo vision
- low resolution
- disparity map
- pairwise
- feature vectors