Reinforcement learning-based design of sampling policies under cost constraints in Markov random fields: Application to weed map reconstruction.
Mathieu BonneauSabrina GabaNathalie PeyrardRégis SabbadinPublished in: Comput. Stat. Data Anal. (2014)
Keyphrases
- markov random field
- maximum a posteriori
- reinforcement learning
- maximum a posteriori probability
- image reconstruction
- graph cuts
- maximum a posterior
- random fields
- map estimation
- potential functions
- image restoration
- energy minimization
- maximum likelihood
- energy function
- belief propagation
- learning algorithm
- parameter estimation
- higher order
- image segmentation
- optimal policy
- message passing
- information extraction
- mrf model
- loopy belief propagation
- low level vision
- machine learning
- piecewise constant functions