Exploring the user guidance for more accurate building segmentation from high-resolution remote sensing images

Dinghao Yang, Bin Wang, Weijia Li, Conghui He

Abstract: Highlights•A building segmentation network supporting various types of user pre-annotations.•A new centroid map prediction module and a new boundary correction module.•Better performance than the state-of-the-art on two popular building datasets.•Extensive experiments on different modules, pre-annotations and cross-domain scenes.