The  Weizmann  Institute  of  Science
                  Faculty of Mathematics and Computer Science
                          Vision and Robotics Seminar
                     Lecture Hall, Room 1, Ziskind Building
                            on Monday, June 6, 2011
                                    at 13:00
                             Note unusual day/time
                                Stephane Mallat
                           CMAP, Ecole Polytechnique
                                 will speak on
              Image Classification with Invariant Representations
Classification requires finding appropriate metrics to compare signals.  Visual
perceptual metrics remain a mystery. They are invariant to particular groups of
transformations such as translations and scaling and are continuous to
deformations relatively to these group actions.  Such invariant representations
can linearize classification problems, but finding them is a challenge.
We introduce a class of non-linear scattering operators which satisfy
invariance and deformation continuity properties. They are computed by
cascading modulus operators and wavelets transforms on the group. For
translations, it provides new representations of stationary processes which
discriminates textures having same power spectrum. A linear PCA classification
applied to scattering representation is shown to provide state of the art
classification results on data bases of structured patterns and for textures.
Relations of invariant scattering with vision and auditory physiology will be
Technion Math Net-2 (TECHMATH2)
Editor: Michael Cwikel   <> 
Announcement from: Diana Mandelik   <>