Fractional Laplacian Pyramids

We provide an extension of the L-2-spline pyramid (Unser et al., 1993) using polyharmonic splines. We analytically prove that the corresponding error pyramid behaves exactly as a multi-scale Laplace operator. We use the multiresolution properties of polyharmonic splines to derive an efficient, non-separable filterbank implementation. Finally, we illustrate the potentials of our pyramid by performing an estimation of the parameters of multivariate fractal processes.

Graph Chatbot

Domain adaptation via alignment of operation profile for Remaining Useful Lifetime prediction

Safe Deep Neural Networks

Statistical Inference for Inverse Problems: From Sparsity-Based Methods to Neural Networks

Domain adaptation via alignment of operation profile for Remaining Useful Lifetime prediction

Safe Deep Neural Networks

Statistical Inference for Inverse Problems: From Sparsity-Based Methods to Neural Networks