Plan2Map: A Multimodal Benchmark for Document-Grounded Geospatial Boundary Reconstruction from Planning Records
arXiv:2606.02747v1 Announce Type: cross Abstract: Planning records define restrictions over geographic areas, but their source documents often provide only indirect spatial evidence rather than machine-readable boundaries. We introduce Plan2Map, a 208-case multimodal benchmark for document-grounded geospatial boundary reconstruction from UK planning records. Given only a source planning document, systems must reconstruct a valid geospatial boundary from notice text, schedules, map plates, map labels, and boundary annotations; the reference GeoJSON is held out for scoring. We propose GeoPlanAge
The proliferation of AI models capable of multimodal understanding and reasoning makes automated geospatial boundary reconstruction from diverse, unstructured planning documents newly feasible.
Accurate, automated geospatial boundary reconstruction can significantly reduce the costs and time associated with land management, urban planning, and infrastructure development, unlocking vast troves of previously inaccessible data.
This benchmark signifies a shift towards using AI to interpret complex, unstructured planning records, moving beyond manual data entry and GIS system constraints to directly convert legal and descriptive documents into actionable spatial data.
- · GIS software providers
- · Urban planners
- · Real estate developers
- · Government agencies
- · Manual geospatial data entry services
- · Legacy GIS data consultancies
Automated systems will be able to quickly process and digitize historical and current planning documents, creating comprehensive, machine-readable maps.
Improved spatial data accessibility will streamline regulatory processes, reduce legal disputes over land boundaries, and enable more efficient infrastructure projects.
The ability to rapidly generate detailed geospatial data from planning records could significantly enhance dynamic urban simulations and 'digital twin' initiatives, transforming smart city development.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.AI