dc.contributor.author | Liu, Qinghui | |
dc.contributor.author | Kampffmeyer, Michael | |
dc.contributor.author | Jenssen, Robert | |
dc.contributor.author | Salberg, Arnt-Børre | |
dc.date.accessioned | 2021-03-11T11:18:20Z | |
dc.date.available | 2021-03-11T11:18:20Z | |
dc.date.created | 2021-03-09T11:25:33Z | |
dc.date.issued | 2020 | |
dc.identifier.isbn | 9781728163741 | |
dc.identifier.uri | https://hdl.handle.net/11250/2732845 | |
dc.description.abstract | Graph Neural Networks (GNNs) have received increasing attention in many fields. However, due to the lack of prior graphs, their use for semantic labeling has been limited. Here, we propose a novel architecture called the Self-Constructing Graph (SCG), which makes use of learnable latent variables to generate embeddings and to self-construct the underlying graphs directly from the input features without relying on manually built prior knowledge graphs. SCG can automatically obtain optimized non-local context graphs from complex-shaped objects in aerial imagery. We optimize SCG via an adaptive diagonal enhancement method and a variational lower bound that consists of a customized graph reconstruction term and a Kullback-Leibler divergence regularization term. We demonstrate the effectiveness and flexibility of the proposed SCG on the publicly available ISPRS Vaihingen dataset and our model SCG-Net achieves competitive results in terms of F1-score with much fewer parameters and at a lower computational cost compared to related pure-CNN based work. | |
dc.language.iso | eng | en_US |
dc.relation.ispartof | IGARSS 2020 - 2020 IEEE International Geoscience and Remote Sensing Symposium. Proceedings | |
dc.rights | Navngivelse-Ikkekommersiell-DelPåSammeVilkår 4.0 Internasjonal | * |
dc.rights.uri | http://creativecommons.org/licenses/by-nc-sa/4.0/deed.no | * |
dc.title | Self-Constructing Graph Convolutional Networks for Semantic Labeling | en_US |
dc.type | Chapter | en_US |
dc.description.version | submittedVersion | |
cristin.ispublished | true | |
cristin.fulltext | preprint | |
cristin.qualitycode | 1 | |
dc.identifier.cristin | 1896600 | |
dc.relation.project | Norges forskningsråd: 272399 | |