Classification with invariant scattering representations

Joan Bruna, Stéphane Mallat

Research output: Chapter in Book/Report/Conference proceedingConference contribution


A scattering transform defines a signal representation which is invariant to translations and Lipschitz continuous relatively to deformations. It is implemented with a non-linear convolution network that iterates over wavelet and modulus operators. Lipschitz continuity locally linearizes deformations. Complex classes of signals and textures can be modeled with low-dimensional affine spaces, computed with a PCA in the scattering domain. Classification is performed with a penalized model selection. State of the art results are obtained for handwritten digit recognition over small training sets, and for texture classification.

Original languageEnglish (US)
Title of host publication2011 IEEE 10th IVMSP Workshop
Subtitle of host publicationPerception and Visual Signal Analysis, IVMSP 2011 - Proceedings
PublisherIEEE Computer Society
Number of pages6
ISBN (Print)9781457712869
StatePublished - 2011

Publication series

Name2011 IEEE 10th IVMSP Workshop: Perception and Visual Signal Analysis, IVMSP 2011 - Proceedings


  • Image classification
  • Invariant representations
  • local image descriptors
  • pattern recognition
  • texture classification

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition
  • Signal Processing


Dive into the research topics of 'Classification with invariant scattering representations'. Together they form a unique fingerprint.

Cite this