Loading
Research · Thread 08
Floor-plan extraction
Six pipeline generations, scored on the same eighteen scans, and the honest result: the simplest one won.
A scan already knows the shape of a building, but a buyer wants a floor plan: walls, rooms and doors, drawn to scale on one page. The hard part is deciding what counts as a wall, when wardrobes, curtains, kitchen islands and half-height partitions all read as vertical surfaces to a point cloud. Six pipeline generations were built, from fitting geometry directly to the point cloud, through grid and footprint methods, to a learned model built on published architectural-reconstruction research, and every one was scored against the same eighteen scans rather than demoed on the ones it handled well.
Paper to shipped
Implementation
Six generations, kept side by sideEach new generation stayed runnable rather than replacing the last, geometry-fitting through grid and footprint methods to a learned model, which is what made it possible to find out later that a newer one was worse.
Benchmark
The learned model lost, and kept losingThe machine-learning pipeline is the most sophisticated approach here and wins on two scans out of eighteen. It was not dismissed on a first result: the training scans were auto-labelled, it went through two rounds of fine-tuning, the scoring was overhauled, and the output polygons were expanded. Its scores improved substantially; the win rate did not move.
In progress
Why it loses, and what ships insteadIt learns from synthetic buildings and is asked to draw real scanned ones, coming out 30 to 80 percent smaller than the true area, a gap between training data and deployment rather than a tuning problem. The plans published on this site today are drawn by the ninth version, a plain grid method, labelled grid_v9 in its own title block. Next: pairing the learned model with something that gets the outer shape right.