Optical flow based Kalman filter for body joint prediction and tracking using HOG-LBP matching.
Binu M. NairKimberly D. KendricksVijayan K. AsariRonald F. TuttlePublished in: Video Surveillance and Transportation Imaging Applications (2014)
Keyphrases
- high dimensional
- kalman filter
- optical flow
- kalman filtering
- motion prediction
- motion parameters
- particle filter
- target tracking
- object tracking
- robust tracking
- particle filtering
- mean shift
- extended kalman filter
- image sequences
- state estimation
- motion model
- data points
- local binary pattern
- bayesian filtering
- unscented kalman filter
- tracking framework
- action recognition
- state space model
- data association
- object detection
- histograms of oriented gradients
- face recognition
- feature descriptors
- moving objects
- sift descriptors
- feature tracking
- image processing
- human detection
- flow field
- background subtraction
- human body
- multi frame
- feature matching
- visual tracking
- image matching
- feature vectors
- object recognition
- feature extraction
- computer vision
- update equations