Riesz Networks: Scale-Invariant Neural Networks in a Single Forward Pass
Open Access
- 29 February 2024
- journal article
- research article
- Published by Springer Nature in Journal of Mathematical Imaging and Vision
- Vol. 66 (3) , 246-270
Abstract
Scale invariance of an algorithm refers to its ability to treat objects equally independently of their size. For neural networks, scale invariance is typically achieved by data augmentation. However, when presented with a scale far outside the range covered by the training set, neural networks may fail to generalize. Here, we introduce the Riesz network, a novel scale- invariant neural network. Instead of standard 2d or 3d convolutions for combining spatial information, the Riesz network is based on the Riesz transform which is a scale-equivariant operation. As a consequence, this network naturally generalizes to unseen or even arbitrary scales in a single forward pass. As an application example, we consider detecting and segmenting cracks in tomographic images of concrete. In this context, ‘scale’ refers to the crack thickness which may vary strongly even within the same sample. To prove its scale invariance, the Riesz network is trained on one fixed crack width. We then validate its performance in segmenting simulated and real tomographic images featuring a wide range of crack widths. An additional experiment is carried out on the MNIST Large Scale data set.Keywords
All Related Versions
Funding Information
- Rheinland-Pfälzische Technische Universität Kaiserslautern-Landau
This publication has 42 references indexed in Scilit:
- Scale Equivariant Neural Networks with Morphological Scale-Spaces2021
- Influence of specimen size on the fibre geometry and tensile strength of ultra‐high‐performance fibre‐reinforced concreteStructural Concrete, 2021
- Stahlfaserbeton – Charakterisierung von rissquerenden Fasern mittels Computer‐TomografieBeton- und Stahlbetonbau, 2021
- Influence of production parameters on the fiber geometry and the mechanical behavior of ultra high performance fiber‐reinforced concreteStructural Concrete, 2020
- A Unified Multi-scale Deep Convolutional Neural Network for Fast Object Detection2016
- U-Net: Convolutional Networks for Biomedical Image Segmentation2015
- Generalized Analytic Signals in Image Processing: Comparison, Theory and Applications2013
- Differential Geometry of Monogenic Signal Representations2008
- Riesz-Transforms Versus Derivatives: On the Relationship Between the Boundary Tensor and the Energy Tensor2005
- Scale Adaptive Filtering Derived from the Laplace EquationLecture Notes in Computer Science, 2001