Graph-based Isometry Invariant Representation Learning

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.

Cited by in corpus (14)

• A Comprehensive Survey of Graph Embedding: Problems, Techniques and Applications
• Edge Attention-based Multi-Relational Graph Convolutional Networks
• Image Classification with Hierarchical Multigraph Networks
• Desiderata for Representation Learning: A Causal Perspective
• Spectral Graph Attention Network with Fast Eigen-approximation
• Temporal Transformer Networks: Joint Learning of Invariant and Discriminative Time Warping
• Non-Parametric Transformation Networks
• Message Passing in Graph Convolution Networks via Adaptive Filter Banks
• Isometric Transformation Invariant Graph-based Deep Neural Network
• Spherical Convolutional Neural Networks: Stability to Perturbations in SO(3)
• Learning Non-Parametric Invariances from Data with Permanent Random Connectomes
• On variation of gradients of deep neural networks
• Equivariance-bridged SO(2)-Invariant Representation Learning using Graph Convolutional Network
• Graph Pooling with Node Proximity for Hierarchical Representation Learning