Automating Procedural Modeling of Buildings from Point Cloud Data

Newer techniques in data collection such as Lidar and photogrammetry can provide large quantities of accurate and up-to-date source data models in operational areas, but transforming this often massive amount of raw source data into a lightweight 3D representation that can be quickly consumed by defense customers using a web browser or mobile devices remains a challenging problem. While point cloud data is increasing in availability, collection frequency, volume and density, the ability to efficiently transmit a meaningful interactive 3D representation of the data remains constrained by connection availability, reliability, and bandwidth. This research task aims to identify a standards-based system to rapidly produce a simplified representation of input sensor data and make it available at the point of need while respecting the connectivity constraints of the end user.One solution is to utilize procedurally generated models—models that are recreated from key data attributes extracted from the source point cloud data. Such models can be generated from procedural rules that utilize extracted point cloud feature attributes as data inputs to guide the building generation process, allowing for reproduction of the original building in a geometric form that requires substantially less processing power, bandwidth, and memory on an end-user platform.