Riesz Networks: Scale-Invariant Neural Networks in a Single Forward Pass

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.

VoroCrack3d: An annotated semi-synthetic 3d image data set of cracked concrete

“…This makes them predestined to be used for training machine learning models. Trained on semi-synthetic images, these models were already successfully applied in many contexts such as crack segmentation in concrete [2 , [13] , [14] , [15] and defect segmentation on metal surfaces [12] . Furthermore, segmentation methods – both from classical image processing and machine learning – can be validated objectively [2 , 3] .…”