Edge-Preserving Bayesian Image Superresolution Based on Compound Markov Random Fields.
Atsunori KanemuraShin-ichi MaedaShin IshiiPublished in: ICANN (2) (2007)
Keyphrases
- edge preserving
- image restoration
- image prior
- markov random field
- super resolution
- gauss markov random fields
- degraded images
- image super resolution
- high resolution
- maximum a posteriori
- low resolution images
- random fields
- image segmentation
- low resolution
- super resolved
- total variation
- image reconstruction
- map estimation
- energy function
- mrf model
- prior model
- high resolution images
- noise reduction
- anisotropic diffusion
- graph cuts
- super resolution reconstruction
- image processing
- super resolution algorithm
- multiscale
- textured images
- parameter estimation
- higher order
- least squares
- spatial resolution
- blurred images
- multi frame
- energy minimization
- input image
- mrf models
- motion blur
- maximum likelihood
- edge detection
- image analysis
- blind deconvolution
- optical flow
- nonlinear diffusion
- bayesian estimation
- bayesian framework
- potential functions