Benchmarking the Future: Automated 2D Floor Plan Generation from Reality Capture

This panel will report on recent USIBD-led efforts to benchmark, validate, and advance emerging methods that automatically generate accurate 2D floor plans directly from 3D reality mapping data captured through laser scanning and close-range photogrammetry. As the demand for digital twins, building documentation, and real estate visualization accelerates, automation has become essential. Traditional workflows for converting 3D scans into floor plans remain time-consuming, labor-intensive, and prone to human interpretation and error. Through an interactive discussion, panelists will present findings and insights from the Reality Mapping to Floor Plan Challenge, highlighting performance trends, methodological breakthroughs, and gaps revealed by benchmarking activities conducted across industry and academic solutions. The conversation will explore how automated approaches can dramatically reduce manual drafting effort while improving accuracy, repeatability, and consistency across the built-environment ecosystem. This session aims to catalyze innovation at the intersection of reality capture, artificial intelligence, and building documentation—inviting practitioners, researchers, and solution developers to collectively shape the next generation of floor plan extraction technologies and standards.