GBSS:a global building semantic segmentation dataset for large-scale remote sensing building extraction
This dataset addresses the problem of limited diversity in training data for remote sensing building extraction, which is incremental as it builds upon existing datasets by providing more varied samples.
The authors tackled the need for diverse training samples in large-scale building extraction by constructing the Global Building Semantic Segmentation (GBSS) dataset, which includes 116.9k sample pairs (about 742k buildings) from six continents and serves as a challenging benchmark for evaluating model generalization and robustness.
Semantic segmentation techniques for extracting building footprints from high-resolution remote sensing images have been widely used in many fields such as urban planning. However, large-scale building extraction demands higher diversity in training samples. In this paper, we construct a Global Building Semantic Segmentation (GBSS) dataset (The dataset will be released), which comprises 116.9k pairs of samples (about 742k buildings) from six continents. There are significant variations of building samples in terms of size and style, so the dataset can be a more challenging benchmark for evaluating the generalization and robustness of building semantic segmentation models. We validated through quantitative and qualitative comparisons between different datasets, and further confirmed the potential application in the field of transfer learning by conducting experiments on subsets.