Triggered by Sebastien Bubeck blog entry posted on the Google+ group, I got to find out that Mallat et al 's scattering transform is now hosted under a toolbox called ScatNet. From the page:
ScatNet is a MATLAB implementation of Scattering Networks transforms and classification algorithms, with reproduction of experiments and figures from papers. It includes:
- Scattering of one-dimensional signals  (figures reproduced)
- Time scattering of audio signals  (figures reproduced)
- Frequency Scattering for audio  (figures reproduced)
- Scattering for images [2, 4]
- Roto-translation scattering for images References : Mathematical introduction of scattering operators for translation and rotation invariant representations, S. Mallat, "Group Invariant Scattering" Communications in Pure and Applied Mathematics, October 2012. Scattering transform for image classification, J.Bruna and S. Mallat, IEEE Trans. on PAMI, August 2013 : "Invariant Scattering Convolution Network" Scattering for audio signals, J. Anden, S. Mallat "Deep Scattering Spectrum" , http://arxiv.org/abs/1304.6763, submitted to IEEE trans on signal processing. Rotation invariant scattering for images, L. Sifre and S. Mallat, "Rotation, Scaling and Deformation Invariant Scattering for Texture Discrimination", in Proc. IEEE CVPR 2013 conference.
Liked this entry ? subscribe to Nuit Blanche's feed, there's more where that came from. You can also subscribe to Nuit Blanche by Email, explore the Big Picture in Compressive Sensing or the Matrix Factorization Jungle and join the conversations on compressive sensing, advanced matrix factorization and calibration issues on Linkedin.