Action-Driven Visual Object Tracking With Deep Reinforcement Learning.
Sangdoo YunJongwon ChoiYoungjoon YooKimin YunJin Young ChoiPublished in: IEEE Trans. Neural Networks Learn. Syst. (2018)
Keyphrases
- visual object tracking
- reinforcement learning
- action selection
- mean shift
- object tracking
- partially observable domains
- state action
- function approximation
- action space
- reward shaping
- transition model
- optimal policy
- state space
- learning algorithm
- markov decision processes
- reinforcement learning algorithms
- sensory inputs
- agent learns
- active learning
- support vector
- fitted q iteration
- agent receives
- data sets
- function approximators
- multiple instance learning
- graph cuts
- particle filter
- dynamic programming
- learning process
- high dimensional
- feature extraction
- machine learning