Abstract
This disclosure describes systems and methods for generating geometric details for digital maps. Some existing cartographic rendering systems may rely on interpolation from schematic data, which can result in visual inaccuracies when representing certain geographic features. The described technology can utilize generative artificial intelligence models to generate geometric primitives, such as polygons and polylines, that represent fine-grained map details. These models can be conditioned on both semantic map data and real-world geospatial data, for example, aerial or ground-level imagery. By generating vector graphics for map features and storing them in a discrete presentation layer, this approach can facilitate the creation of map representations with increased visual detail and fidelity for applications such as navigation and logistics planning.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Isert, Carsten; Bertrand Cabral, Manuel; Lynen, Simon; Strosnick, Matthew; and Pallone, Michelina Mary Galligan, "Generating Geospatial Visualizations Using Generative Artificial Intelligence", Technical Disclosure Commons, ()
https://www.tdcommons.org/dpubs_series/10104