Gradient flow on extensive-rank positive semi-definite matrix denoising.
Antoine BodinNicolas MacrisPublished in: ITW (2023)
Keyphrases
- positive semi definite
- denoising
- gradient flow
- image denoising
- partial differential equations
- dot product
- similarity function
- metric learning
- kernel methods
- kernel function
- energy functional
- noisy images
- total variation
- active contours
- natural images
- perceptron algorithm
- level set
- image processing
- curve evolution
- variational formulation
- active contour model
- kernel matrix
- computer vision
- gaussian noise
- pairwise
- loss function
- noise removal
- distance metric
- reproducing kernel hilbert space
- anisotropic diffusion
- multi class
- feature space
- support vector