FloorplanVLM: A Vision-Language Model for Floorplan Vectorization

TL;DR

FloorplanVLM introduces a novel vision-language model that converts raster floorplans into structured vector graphics using sequence modeling, achieving high accuracy and generalization on complex layouts.

Contribution

The paper presents a unified 'pixels-to-sequence' approach for floorplan vectorization, along with a large-scale dataset and a new benchmark for evaluation.

Findings

Achieves 92.52% external-wall IoU on benchmark

Demonstrates robust generalization to non-Manhattan layouts

Outperforms pixel-based and query-based methods

Abstract

Converting raster floorplans into engineering-grade vector graphics is challenging due to complex topology and strict geometric constraints. To address this, we present FloorplanVLM, a unified framework that reformulates floorplan vectorization as an image-conditioned sequence modeling task. Unlike pixel-based methods that rely on fragile heuristics or query-based transformers that generate fragmented rooms, our model directly outputs structured JSON sequences representing the global topology. This 'pixels-to-sequence' paradigm enables the precise and holistic constraint satisfaction of complex geometries, such as slanted walls and curved arcs. To support this data-hungry approach, we introduce a scalable data engine: we construct a large-scale dataset (Floorplan-2M) and a high-fidelity subset (Floorplan-HQ-300K) to balance geometric diversity and pixel-level precision. We then employ a progressive training strategy, using Supervised Fine-Tuning (SFT) for structural grounding and quality annealing, followed by Group Relative Policy Optimization (GRPO) for strict geometric alignment. To standardize evaluation on complex layouts, we establish and open-source FPBench-2K. Evaluated on this rigorous benchmark, FloorplanVLM demonstrates exceptional structural validity, achieving $\textbf{92.52%}$ external-wall IoU and robust generalization across non-Manhattan architectures.