A Method for Visual Model Learning During Tracking.
Miranda GrahlFrank JoublinFranz KummertPublished in: Aust. J. Intell. Inf. Process. Syst. (2010)
Keyphrases
- mathematical model
- prior knowledge
- theoretical analysis
- learning scheme
- cost function
- probabilistic model
- learning mechanism
- statistical model
- test data
- objective function
- prediction model
- neural nets
- energy function
- parameter estimation
- kalman filter
- high level
- classification method
- linear regression
- bp neural network
- evaluation method
- linear model
- modeling method
- linear predictors
- similarity measure
- computational model
- unsupervised learning
- optimization method
- bayesian framework
- em algorithm
- tracking framework
- learned models
- learning process
- multiple kernel
- object tracking
- neural network
- appearance information
- reinforcement learning
- object detectors
- bayesian methods
- video sequences
- autoregressive
- pairwise
- active learning
- input data
- maximum likelihood
- learning tasks
- appearance model