Data Fusion: Maintaining 3D Data at Scale (Sponsored by Ordnance Survey)

3D Semantic Segmentation through deep learning marks a step towards automated data processing. Although the past decade has seen improvements in performance alongside increasingly complex architectures, challenges remain. In partnership with Ordnance Survey (OS), this project aims to develop a multi-branch pipeline to handle variations in point cloud density, assess the viability of OS 2D contextual information for providing semantic labels to 3D data, and evaluate the impact of point feature inclusion on deep learning model performance. This project will support OS in efficiently managing and updating 3D data by introducing scalable methods and integrating historical and model-based geospatial data.