How Marginal Likelihood Inference Unifies Entropy, Correlation and SNR-Based Stopping in Nonlinear Diffusion Scale-Spaces.
Ramunas GirdziusasJorma LaaksonenPublished in: ACCV (1) (2007)
Keyphrases
- scale spaces
- nonlinear diffusion
- scale space
- signal to noise ratio
- noise reduction
- gaussian process
- model selection
- closed form
- partial differential equations
- bayesian networks
- anisotropic diffusion
- bayesian inference
- vector valued
- diffusion equation
- pattern recognition
- denoising
- feature vectors
- hyperparameters
- speckle noise
- natural images
- image features