FloorplanMAE:A self-supervised framework for complete floorplan generation from partial inputs

2025-06-10

AI Summary

Architectural design requires rapid generation of complete floor plans from incomplete hand-drawn sketches, yet existing methods struggle with generalization to early-stage, low-fidelity sketches. Method: This paper proposes an end-to-end floor plan completion framework based on Masked Autoencoders (MAE), the first to apply MAE to architectural floor plan reconstruction. We introduce FloorplanNet—a dedicated dataset—and design a lightweight Vision Transformer coupled with a structured representation mechanism to enhance robustness to real-world sketch variability. Our approach combines self-supervised pretraining with joint geometric-functional optimization. Contribution/Results: The method achieves state-of-the-art performance across layout completeness, functional coherence, and geometric plausibility. Experiments demonstrate its effectiveness in reconstructing authentic early-design sketches into high-quality, editable floor plans—significantly improving conceptual design efficiency and responsiveness.

Technology Category

• Machine Learning: Deep Neural Architectures and Foundation Models
• Planning, Routing, and Scheduling: Planning with Language Models
• Cognitive Modeling & Cognitive Systems: Agent Architectures

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Economics, Online Markets and Human Computation: Architectures and workflows that use LLMs for crowd work
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

In the architectural design process, floorplan design is often a dynamic and iterative process. Architects progressively draw various parts of the floorplan according to their ideas and requirements, continuously adjusting and refining throughout the design process. Therefore, the ability to predict a complete floorplan from a partial one holds significant value in the design process. Such prediction can help architects quickly generate preliminary designs, improve design efficiency, and reduce the workload associated with repeated modifications. To address this need, we propose FloorplanMAE, a self-supervised learning framework for restoring incomplete floor plans into complete ones. First, we developed a floor plan reconstruction dataset, FloorplanNet, specifically trained on architectural floor plans. Secondly, we propose a floor plan reconstruction method based on Masked Autoencoders (MAE), which reconstructs missing parts by masking sections of the floor plan and training a lightweight Vision Transformer (ViT). We evaluated the reconstruction accuracy of FloorplanMAE and compared it with state-of-the-art benchmarks. Additionally, we validated the model using real sketches from the early stages of architectural design. Experimental results show that the FloorplanMAE model can generate high-quality complete floor plans from incomplete partial plans. This framework provides a scalable solution for floor plan generation, with broad application prospects.

Problem

• Predict complete floorplans from partial inputs
• Improve design efficiency with automated generation
• Reduce workload in iterative architectural design

Innovation

• Self-supervised learning for floorplan completion
• Masked Autoencoders reconstruct missing floorplan parts
• Lightweight Vision Transformer trained on FloorplanNet dataset