Human Motion Tracking with Monocular Video by Introducing a Graph Structure into Gaussian Process Dynamical Models.
Jianfeng XuKoichi TakagiShigeyuki SakazawaPublished in: PSIVT (1) (2011)
Keyphrases
- graph structure
- gaussian process
- dynamical models
- dynamical model
- gaussian processes
- video sequences
- pose tracking
- image sequences
- graphical models
- model selection
- bayesian framework
- latent variables
- regression model
- semi supervised
- video frames
- dynamical systems
- directed graph
- visual tracking
- human motion
- hyperparameters
- pose estimation
- spatial and temporal
- probabilistic model
- particle filter
- motion tracking
- motion analysis
- three dimensional
- high dimensional
- active learning
- human pose
- dynamic programming
- motion estimation
- action recognition
- markov random field
- closed form
- learning algorithm
- motion model