Global Building Semantic Segmentation (GBSS) dataset
收藏资源简介:
全球建筑语义分割(GBSS)数据集是由武汉大学遥感信息工程学院创建,包含约742,000个建筑实例,覆盖六大洲,总面积约1310平方公里。数据集通过结合OpenStreetMap和Google Maps的卫星图像,采用半自动化方法构建,旨在提供丰富的样本多样性和广泛的地理覆盖,以支持大规模建筑提取任务。该数据集特别适用于评估模型的泛化能力和鲁棒性,并可用于转移学习研究。
The Global Building Semantic Segmentation (GBSS) Dataset was created by the School of Remote Sensing Information Engineering, Wuhan University. It contains approximately 742,000 building instances, covers six continents, and spans a total area of about 1,310 square kilometers. Developed through a semi-automated approach that combines satellite imagery from OpenStreetMap and Google Maps, this dataset aims to provide rich sample diversity and extensive geographic coverage to support large-scale building extraction tasks. It is particularly suitable for evaluating the generalization ability and robustness of models, and can be used for transfer learning research.

- 1GBSS:a global building semantic segmentation dataset for large-scale remote sensing building extraction武汉大学遥感信息工程学院 · 2024年



