Provably scale-covariant networks from oriented quasi quadrature measures in cascade

This article presents a continuous model for hierarchical networks based on a combination of mathematically derived models of receptive fields and biologically inspired computations. Based on a functional model of complex cells in terms of an oriented quasi quadrature combination of first- and second-order directional Gaussian derivatives, we couple such primitive computations in cascade over combinatorial expansions over image orientations. Scale-space properties of the computational primitives are analysed and it is shown that the resulting representation allows for provable scale and rotation covariance. A prototype application to texture analysis is developed and it is demonstrated that a simplified mean-reduced representation of the resulting QuasiQuadNet leads to promising experimental results on three texture datasets.

• Publication year

  2019

• Venue

  Scale Space and Variational Methods in Computer Vision

• Publication date

  2019-03-01

• Fields of study

  Computer Science

• Identifiers
  DOI 10.1007/978-3-030-22368-7_26arXiv 1903.00289
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Describing complex cells in primary visual cortex: a comparison of context and multifilter LN models

  2018cited by this paper
• BRINT: BINARY ROTATION INVARIANT AND NOISE TOLERANT TEXTURE CLASSIFICATION

  2017cited by this paper
• A Spatiotemporal Oriented Energy Network for Dynamic Texture Recognition

  2017cited by this paper
• Dense scale selection over space, time and space-time

  2017cited by this paper
• Local binary features for texture classification: Taxonomy and experimental study

  2017cited by this paper
• Feature Pyramid Networks for Object Detection

  2016cited by this paper
• Median Robust Extended Local Binary Pattern for Texture Classification

  2016cited by this paper
• A Unified Multi-scale Deep Convolutional Neural Network for Fast Object Detection

  2016cited by this paper
• Using goal-driven deep learning models to understand sensory cortex

  2016cited by this paper
• Spatial Transformer Networks

  2015cited by this paper
• Deep filter banks for texture recognition and segmentation

  2015influential reference
• Origin and Function of Tuning Diversity in Macaque Visual Cortex.

  2015cited by this paper
• Deep Residual Learning for Image Recognition

  2015cited by this paper
• Very Deep Convolutional Networks for Large-Scale Image Recognition

  2014cited by this paper
• PCANet: A Simple Deep Learning Baseline for Image Classification?

  2014cited by this paper
• BRINT: Binary Rotation Invariant and Noise Tolerant Texture Classification

  2014cited by this paper
• A computational theory of visual receptive fields

  2013cited by this paper
• ImageNet classification with deep convolutional neural networks

  2012cited by this paper
• Invariant Scattering Convolution Networks

  2012influential reference
• Multi-dimensional local binary pattern descriptors for improved texture analysis

  2012cited by this paper
• Generalized Gaussian Scale-Space Axiomatics Comprising Linear Scale-Space, Affine Scale-Space and Spatio-Temporal Scale-Space

  2011cited by this paper
• Normalization as a canonical neural computation

  2011cited by this paper
• A new texture descriptor using multifractal analysis in multi-orientation wavelet pyramid

  2010cited by this paper
• A Statistical Approach to Material Classification Using Image Patch Exemplars

  2009cited by this paper
• The Orientation Selectivity of Color-Responsive Neurons in Macaque V1

  2008cited by this paper
• Speeded-Up Robust Features (SURF)

  2008cited by this paper
• Robust Object Recognition with Cortex-Like Mechanisms

  2007cited by this paper
• Spatial structure of complex cell receptive fields measured with natural images.

  2005cited by this paper
• Multiresolution Gray-Scale and Rotation Invariant Texture Classification with Local Binary Patterns

  2002cited by this paper
• Spatial and temporal receptive fields of geniculate and cortical cells and directional selectivity.

  2000cited by this paper
• Feature Detection with Automatic Scale Selection

  1998cited by this paper
• Scale-Space Theory in Computer Vision

  1993cited by this paper
• Normalization of cell responses in cat striate cortex

  1992cited by this paper
• Generic Neighborhood Operators

  1992cited by this paper
• Receptive field families

  1990cited by this paper
• Spatiotemporal energy models for the perception of motion.

  1985cited by this paper
• Neocognitron: A self-organizing neural network model for a mechanism of pattern recognition unaffected by shift in position

  1980cited by this paper

• Scale-Covariant and Scale-Invariant Gaussian Derivative Networks

  2020cites this paper
• Provably Scale-Covariant Continuous Hierarchical Networks Based on Scale-Normalized Differential Expressions Coupled in Cascade

  2019cites this paper
• Provably scale-covariant hierarchical continuous networks based on scale-normalized differential expressions coupled in cascade

  2019cites this paper