Matlab Toolbox for Shearlets
ShearLab is a MATLAB library designed for processing two- and three-dimensional data with a specific class of basis functions called shearlets . Such shearlet systems are particularly well suited to represent anisotropic features (such as curves), which are often crucial in multidimensional data. The resulting representation has proven to be well suited for image processing tasks such as inpainting, denoising or image separation. On the website below we provide the full MATLAB code, a framework for numerical tests and general information about shearlets.
The ShearLab 3D Toolbox (Ver. 1.0) and older versions of ShearLab can be downloaded from http://www.shearlab.org. You can also find ShearLab3D on the Oberwolfach References on Mathematical Software website here and on swMATH here.
This program is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
If you have any questions or comments, please contact Dr. Hector Andrade Loarca (personal website)
CartoonX
The CartoonX method was presented in Cartoon Explanation of Image Classifiers - S.Kolek. et al. ECCV 2022 (oral presentation) as an explanation method for black-box image classifiers such as deep neural networks. CartoonX extracts the piecewise smooth relevant part of an image by optimizing an optimization mask in the wavelet domain of an image. The code repository for the CartoonX method (implemented in PyTorch) can be found at https://github.com/skmda37/CartoonX.
If you have any questions, please contact Stefan Kolek.