@InProceedings{AvelarTavSilJunLam:2020:SuImCl,
               author = "Avelar, Pedro Henrique da Costa and Tavares, Anderson Rocha and 
                         Silveira, Thiago Lopes Trugillo da and Jung, Cl{\'a}udio Rosito 
                         and Lamb, Lu{\'{\i}}s da Cunha",
          affiliation = "{Federal University of Rio Grande do Sul} and {Federal University 
                         of Rio Grande do Sul} and {University of Rio Grande} and {Federal 
                         University of Rio Grande do Sul} and {Federal University of Rio 
                         Grande do Sul}",
                title = "Superpixel Image Classification with Graph Attention Networks",
            booktitle = "Proceedings...",
                 year = "2020",
               editor = "Musse, Soraia Raupp and Cesar Junior, Roberto Marcondes and 
                         Pelechano, Nuria and Wang, Zhangyang (Atlas)",
         organization = "Conference on Graphics, Patterns and Images, 33. (SIBGRAPI)",
            publisher = "IEEE Computer Society",
              address = "Los Alamitos",
             keywords = "superpixel,graph attention networks,graph neural networks.",
             abstract = "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.",
  conference-location = "Porto de Galinhas (virtual)",
      conference-year = "7-10 Nov. 2020",
                  doi = "10.1109/SIBGRAPI51738.2020.00035",
                  url = "http://dx.doi.org/10.1109/SIBGRAPI51738.2020.00035",
             language = "en",
                  ibi = "8JMKD3MGPEW34M/43BDF3B",
                  url = "http://urlib.net/ibi/8JMKD3MGPEW34M/43BDF3B",
           targetfile = "PID6630943.pdf",
        urlaccessdate = "2025, Jan. 15"
}