Adaptive visual tracking using the prioritized Q-learning algorithm: MDP-based parameter learning approach.
Sarang KhimSungjin HongYoonyoung KimPhill-Kyu RheePublished in: Image Vis. Comput. (2014)
Keyphrases
- visual tracking
- parameter learning
- learning algorithm
- particle filter
- particle filtering
- structure learning
- appearance model
- generative model
- conditional random fields
- bayesian networks
- reinforcement learning
- mean shift
- statistical learning
- video sequences
- machine learning algorithms
- maximum likelihood
- markov random field
- em algorithm
- object tracking
- active learning
- state space
- training data
- computer vision
- probabilistic model
- random variables
- expectation maximization
- higher order
- supervised learning
- approximate inference
- optical flow
- machine learning