Semantic Segmentation on Swiss3DCities: A Benchmark Study on Aerial Photogrammetric 3D Pointcloud Dataset
This dataset provides a new benchmark for semantic segmentation in urban outdoor environments, which is useful for researchers and developers in autonomous driving, gaming, and smart city planning.
This paper introduces Swiss3DCities, a new 2.7 km^2 outdoor urban 3D pointcloud dataset from three Swiss cities, manually annotated for semantic segmentation. The dataset is built using photogrammetry from multirotor-acquired images, resulting in uniformly dense and complete point clouds. The authors benchmark PointNet++ on this dataset and study the impact of using different cities for model generalization.
We introduce a new outdoor urban 3D pointcloud dataset, covering a total area of 2.7 $km^2$, sampled from three Swiss cities with different characteristics. The dataset is manually annotated for semantic segmentation with per-point labels, and is built using photogrammetry from images acquired by multirotors equipped with high-resolution cameras. In contrast to datasets acquired with ground LiDAR sensors, the resulting point clouds are uniformly dense and complete, and are useful to disparate applications, including autonomous driving, gaming and smart city planning. As a benchmark, we report quantitative results of PointNet++, an established point-based deep 3D semantic segmentation model; on this model, we additionally study the impact of using different cities for model generalization.