An Annotated Multi-Platform Laser Scanning Dataset (MLS, ULS, TLS) for Semantic Segmentation of Point Clouds in Complex Forest Environments

Available datasets for semantic segmentation of forest point clouds mostly contain data from a single sensing system. This dataset provides LiDAR point clouds of a forest scene in the Richterschlüchte area, Saxon Switzerland, Germany, acquired during a field campaign in March 2023 using three different laser scanning systems: mobile laser scanning (MLS), uncrewed aerial vehicle laser scanning (ULS), and terrestrial laser scanning (TLS). All point clouds were manually annotated with three semantic classes, namely trunk, ground, and miscellaneous. In total, five point clouds are included: the MLS point cloud is partitioned into training, validation, and test subsets, while the TLS and ULS point clouds are provided to assess generalization across varying sensing systems. The dataset is intended for developing and testing deep learning approaches for the semantic segmentation of point clouds in complex forest environments.