Saved in:
Bibliographic Details
Main Authors: Liu, Yuanqing, Yang, Ziming, Li, Yulong, Yang, Yue
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2602.06507
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911427532423168
author Liu, Yuanqing
Yang, Ziming
Li, Yulong
Yang, Yue
author_facet Liu, Yuanqing
Yang, Ziming
Li, Yulong
Yang, Yue
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2602_06507
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle FloorplanVLM: A Vision-Language Model for Floorplan Vectorization
Liu, Yuanqing
Yang, Ziming
Li, Yulong
Yang, Yue
Computer Vision and Pattern Recognition
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.
title FloorplanVLM: A Vision-Language Model for Floorplan Vectorization
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2602.06507