Superpixel Image Classification with Graph Attention Networks
This work addresses image classification for scenarios like panoramas where traditional methods fail, though it is incremental as it adapts existing techniques to a new application.
The paper tackles image classification by transforming images into region adjacency graphs using superpixels and applying Graph Attention Networks, finding that GATs outperform other GNN models but underperform raw image classifiers due to information loss.
This paper presents a methodology for image classification using Graph Neural Network (GNN) models. We transform the input images into region adjacency graphs (RAGs), in which regions are superpixels and edges connect neighboring superpixels. Our experiments suggest that Graph Attention Networks (GATs), which combine graph convolutions with self-attention mechanisms, outperforms other GNN models. Although raw image classifiers perform better than GATs due to information loss during the RAG generation, our methodology opens an interesting avenue of research on deep learning beyond rectangular-gridded images, such as 360-degree field of view panoramas. Traditional convolutional kernels of current state-of-the-art methods cannot handle panoramas, whereas the adapted superpixel algorithms and the resulting region adjacency graphs can naturally feed a GNN, without topology issues.