The differogram: Non-parametric noise variance estimation and its use for model selection.
Kristiaan PelckmansJos De BrabanterJohan A. K. SuykensBart De MoorPublished in: Neurocomputing (2005)
Keyphrases
- model selection
- noise variance
- regularization parameter
- hyperparameters
- parameter estimation
- cross validation
- gaussian process
- selection criterion
- correlation coefficient
- machine learning
- noisy images
- noise level
- mixture model
- motion segmentation
- model selection criteria
- image noise
- sample size
- feature selection
- gaussian noise
- wiener filter
- feature vectors
- high dimensional
- support vector
- data sets