From Kernel Methods to Neural Networks: A Unifying Variational Formulation.
Michael UnserPublished in: CoRR (2022)
Keyphrases
- kernel methods
- variational formulation
- neural network
- energy functional
- optical flow
- kernel function
- level set
- piecewise smooth
- support vector
- object boundaries
- shape prior
- machine learning
- support vector machine
- reproducing kernel hilbert space
- regularization term
- feature space
- intensity distribution
- pattern recognition
- smooth surfaces
- active contours
- energy function
- level set method
- computer vision
- feature extraction
- high dimensional
- input image
- shape model
- image segmentation
- markov random field