HouseTune: Two-Stage Floorplan Generation with LLM Assistance

This paper proposes a two-stage text-to-floorplan generation framework that combines the reasoning capability of Large Language Models (LLMs) with the generative power of diffusion models. In the first stage, we leverage a Chain-of-Thought (CoT) prompting strategy to guide an LLM in generating an initial layout (Layout-Init) from natural language descriptions, which ensures a user-friendly and intuitive design process. However, Layout-Init may lack precise geometric alignment and fine-grained structural details. To address this, the second stage employs a conditional diffusion model to refine Layout-Init into a final floorplan (Layout-Final) that better adheres to physical constraints and user requirements. Unlike prior methods, our approach effectively reduces the difficulty of floorplan generation learning without the need for extensive domain-specific training data. Experimental results demonstrate that our approach achieves state-of-the-art performance across all metrics, which validates its effectiveness in practical home design applications. Our code will be made publicly available. In architectural design, creating floorplans that align with user requirements remains a challenge. Traditional design methods not only rely on specialized knowledge but also require designers to make iterative adjustments to accommodate specific user needs, which makes personalized design challenging. Learning-based models Murali et al. [2017 ], Merrell et al. [2010 ], Sun et al. [2022 ], Luo and Huang [2022] have made efforts to improve the accuracy, interactivity, and efficiency of floorplan generation. Despite these advances, existing approaches have yet to achieve a level of user-friendliness and accuracy that enables ready adoption by ordinary users.