Enhanced sampling schemes for MCMC based blind Bernoulli-Gaussian deconvolution.
Di GeJérôme IdierEric Le CarpentierPublished in: Signal Process. (2011)
Keyphrases
- markov chain monte carlo
- monte carlo
- motion deblurring
- sampling algorithm
- exponential family
- metropolis hastings
- markov chain
- least squares
- posterior distribution
- dirichlet process
- importance sampling
- generative model
- markov chain monte carlo sampling
- parameter estimation
- motion blur
- maximum likelihood
- bayesian inference
- denoising
- blind deconvolution
- particle filter
- posterior probability
- gibbs sampler
- data association
- image deblurring
- image restoration
- approximate inference
- random sampling
- metropolis hastings algorithm
- total variation
- point spread function
- expectation propagation
- proposal distribution
- probabilistic model
- sample size
- image noise
- prior information
- gaussian noise
- blurred images
- bayesian framework
- density function
- particle filtering
- real valued