What's new in ScatNet v0.2 :
- More efficient wavelet transforms with the cascade algorithm : see core/wavelet_factory_2d_pyramid.m
- Frequency transposition invariance : see core/scat_freq.m
- Roto-translation invariance : see core/wavelet_factory_3d_pyramid.m
- Inverse scattering transform for 1D : see reconstruction/
- Selesnick wavelets of compact support : see filters/selesnick_filter_bank_1d.
- Improved display : see display/
- New utility functions for classification : see scatutils/
- A comprehensive documentation PDF file : see doc/
- 15 demos : see demo/
- Standard headers for all functions
- The MATLAB Signal Processing Toolbox is no longer required
ScatNet is a MATLAB library for scattering networks.
See our homepage of the project for documentation, tutorials, and bibliography
http://www.di.ens.fr/data/software/scatnet/
For any enquiry, please write to the brand new e-mail address of our team : scatnet[AT]di[DOT]ens[DOT]fr where [AT] and [DOT] must be replaced by the characters @ and . respectively.
- launch
addpath_scatnetfrom matlab shell
[optional] : add the following two lines to your startup.m file so that matlab does the addpath automatically when it starts:
addpath /path/to/scatnet;
addpath_scatnet;
Use ScatNet with audio signals: http://www.di.ens.fr/data/software/scatnet/quickstart-audio/
Use ScatNet with images: http://www.di.ens.fr/data/software/scatnet/quickstart-image/
Developers : Laurent Sifre, Joakim Andén, Edouard Oyallon, Michel Kapoko, Vincent Lostanlen.
Beta-testers : Irène Waldspurger, Guy Wolf, Matthew Hirn.
Many thanks to : Stéphane Mallat, our team leader ; Ivan Selesnick (NYU), of which the filters/selesnick/ package appears courtesy ; Joan Bruna (NYU) for his help.