Paris-CARLA-3D is a dataset of several dense colored point clouds of outdoor environments built by a mobile LiDAR and camera system. The data are composed of two sets with synthetic data from the open source CARLA simulator (700 million points) and real data acquired in the city of Paris (60 million points), hence the name Paris-CARLA-3D. One of the advantages of this dataset is to have simulated the same LiDAR and camera platform in the open source CARLA simulator as the one used to produce the real data. In addition, manual annotation of the classes using the semantic tags of CARLA was performed on the real data, allowing the testing of transfer methods from the synthetic to the real data. The objective of this dataset is to provide a challenging dataset to evaluate and improve methods on difficult vision tasks for the 3D mapping of outdoor environments: semantic segmentation, instance segmentation, and scene completion. For each task, we describe the evaluation protocol as well as the experiments carried out to establish a baseline.

3D building segmentation is an important research issue in the remote sensing community with relevant applications to urban modeling, cloud-to-cloud and cloud-to-model registration, 3D cartography, virtual reality, cultural heritage documentation, among others. In this paper, we propose automatic, parametric and robust approaches to segment façades from 3D point clouds. Processing is carried out using elevation images and 3D decomposition, and the final result can be reprojected onto the 3D point cloud for visualization or evaluation purposes. Our methods are based on geometrical and geodesic constraints. Parameters are related to urban and architectural constraints. Thus, they can be set up to manage façades of any height, length and elongation. We propose two methods based on façade marker extraction and a third method without markers based on the maximal elongation image. This work is developed in the framework of TerraMobilita project. The performance of our methods is proved in our experiments on TerraMobilita databases using 2D and 3D ground truth annotations.