Graph-based Isometry Invariant Representation Learning

TL;DR

This paper introduces TIGraNet, a graph-based neural network that learns features inherently invariant to isometric transformations like rotation and translation, improving classification robustness on transformed images.

Contribution

The paper proposes a novel graph-based network architecture that replaces traditional convolution and pooling with spectral graph operations to achieve transformation invariance.

Findings

High accuracy on rotated and translated images

Enhanced robustness to geometric transformations

Maintains performance with limited training data

Abstract

Learning transformation invariant representations of visual data is an important problem in computer vision. Deep convolutional networks have demonstrated remarkable results for image and video classification tasks. However, they have achieved only limited success in the classification of images that undergo geometric transformations. In this work we present a novel Transformation Invariant Graph-based Network (TIGraNet), which learns graph-based features that are inherently invariant to isometric transformations such as rotation and translation of input images. In particular, images are represented as signals on graphs, which permits to replace classical convolution and pooling layers in deep networks with graph spectral convolution and dynamic graph pooling layers that together contribute to invariance to isometric transformations. Our experiments show high performance on rotated and translated images from the test set compared to classical architectures that are very sensitive to transformations in the data. The inherent invariance properties of our framework provide key advantages, such as increased resiliency to data variability and sustained performance with limited training sets.