The paper focuses on building extraction, which is crucial in infrastructure development, population management, and geological observations. It compares shallow models to deep learning models using LiDAR data for supervised segmentation. The research found that shallow models outperform deep learning models in Intersection over Union (IoU) scores, especially when using aerial images. However, deep learning models show superior performance in Boundary Intersection over Union (BIoU) scores. The paper concludes that Light Gradient-Boosting Machine (LightGBM) performs better than RF and Extreme Gradient Boosting (XGBoost).
Publication date: 22 Sep 2023
Project Page: Not provided
Paper: https://arxiv.org/pdf/2309.12027
