Bayesian Image Modeling by Means of a Generalized Sparse Prior and Loopy Belief Propagation
スポンサーリンク
概要
- 論文の詳細を見る
Bayesian image modeling is presented based on a generalized sparse prior probability distribution. Our prior includes sparsity in each interaction term between every pair of neighbouring pixels in Markov random fields. A new scheme for hyperparameter estimation is based on the conditional maximization of entropy in our generalized sparse prior. In addition, the criterion used for defining the optimal value for sparseness in interactions is that of the maximization of marginal likelihood. Our practical algorithm is based on loopy belief propagation.
- 2012-11-15
著者
-
Yasuda Muneki
Graduate School of Information Sciences, Tohoku University, Sendai 980-8579, Japan
-
Yasuda Muneki
Graduate School of Information Science, Tohoku University, Sendai 980-8579, Japan
-
Tanaka Kazuyuki
Graduate School of Information Science, Tohoku University, Sendai 980-8579, Japan
-
TITTERINGTON D.
School of Mathematics and Statistics, University of Glasgow, Glasgow G12 8QQ, U.K.
関連論文
- Inverse Problem in Pairwise Markov Random Fields Using Loopy Belief Propagation
- Learning algorithm in restricted Boltzmann machines using Kullback-Leibler importance estimation procedure
- Bayesian Image Modeling by Means of a Generalized Sparse Prior and Loopy Belief Propagation
- Statistical Analysis of Gaussian Image Inpainting Problems
- Learning Algorithm for Boltzmann Machines Using Max-Product Algorithm and Pseudo-Likelihood
- Pseudo Temperature of Observed Data
- Learning Algorithm for Boltzmann Machines Using Max-Product Algorithm and Pseudo-Likelihood
- Mean Field Approximation for Fields of Experts
- Statistical Performance Analysis in Probabilistic Image Processing