Graph Neural Networks in Computer Vision -- Architectures, Datasets and Common Approaches

TL;DR

This paper reviews the application of various Graph Neural Network architectures in computer vision, analyzing datasets and exploring inter-field inspirations to guide future research.

Contribution

It provides a comprehensive overview of GNN architectures, datasets, and cross-disciplinary insights specific to computer vision applications.

Findings

GNN architectures like GAT, GCN, and GRN are increasingly used in computer vision.

The paper summarizes datasets employed in GNN-based computer vision research.

Relations between GNN studies and other fields are identified for future exploration.

Abstract

Graph Neural Networks (GNNs) are a family of graph networks inspired by mechanisms existing between nodes on a graph. In recent years there has been an increased interest in GNN and their derivatives, i.e., Graph Attention Networks (GAT), Graph Convolutional Networks (GCN), and Graph Recurrent Networks (GRN). An increase in their usability in computer vision is also observed. The number of GNN applications in this field continues to expand; it includes video analysis and understanding, action and behavior recognition, computational photography, image and video synthesis from zero or few shots, and many more. This contribution aims to collect papers published about GNN-based approaches towards computer vision. They are described and summarized from three perspectives. Firstly, we investigate the architectures of Graph Neural Networks and their derivatives used in this area to provide accurate and explainable recommendations for the ensuing investigations. As for the other aspect, we also present datasets used in these works. Finally, using graph analysis, we also examine relations between GNN-based studies in computer vision and potential sources of inspiration identified outside of this field.