Learning and Evaluating Markov Random Fields for Natural Images
Year: 2010
Type: dissertation
Author: Uwe Schmidt
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Keywords: Markov random fields, image priors, natural image statistics, score matching, contrastive divergence, image restoration, generative models, maximum likelihood estimation, low-level vision, regularization, maximum a posteriori estimation, MMSE estimation, efficient sampling, learning methods comparison, posterior mean estimation, probabilistic models
Open Access status: closed
OpenAlex ID: W2946366481
