3datakilder

3datakilder is a term used in Danish and Norwegian contexts to denote sources of three-dimensional data that underpin digital representations of objects, scenes, and spaces. In practice, 3D data sources provide the geometry, textures, and metadata needed for modeling, analysis, and visualization across disciplines such as GIS, architecture, gaming, and robotics.

Common types of 3D data include geometry data (point clouds, meshes, and voxel grids), surface textures (image textures, normal maps, and color maps), and semantic or attribute data (material properties, classifications, and metadata). Data may be static or time-dependent, enabling the study of changes in environments or objects over time. Formats and standards vary by use case, with OBJ, STL, and PLY widely used for geometry; LAS/LAZ for lidar-point clouds; GLTF and 3D Tiles for web-based delivery; CityGML for urban models; and appropriate metadata and coordinate reference systems for georeferencing.

Sources of 3D data range from government open data portals, national mapping agencies, research laboratories, and commercial vendors to crowdsourced and university-generated datasets. Licensing and accessibility vary, spanning public domain, open licenses, and commercial agreements, all of which influence how data can be reused or redistributed.

Processing typically involves cleaning, alignment, and registration; conversion to usable formats; meshing or point-cloud processing; and texture mapping, often with level-of-detail techniques for efficient visualization. Applications include urban planning, simulation and training, virtual and augmented reality, gaming, heritage documentation, and robotics.

Challenges in 3datakilder include data quality and scale, interoperability between formats and CRS, licensing and attribution requirements, privacy considerations in dense urban areas, and maintaining accuracy over time.