Boundary-Constrained Diffusion Models for Floorplan Generation: Balancing Realism and Diversity

TL;DR

This paper introduces boundary-constrained diffusion models for floorplan generation, balancing realism and diversity through new metrics and modules, and highlights the trade-offs and dataset dependencies in current generative approaches.

Contribution

It proposes the Diversity Score metric and Boundary Cross-Attention module to improve diversity and boundary adherence in diffusion-based floorplan generation.

Findings

BCA improves boundary adherence significantly.

Prolonged training reduces diversity despite low FID.

Models rely heavily on dataset priors, affecting generalization.

Abstract

Diffusion models have become widely popular for automated floorplan generation, producing highly realistic layouts conditioned on user-defined constraints. However, optimizing for perceptual metrics such as the Fr\'echet Inception Distance (FID) causes limited design diversity. To address this, we propose the Diversity Score (DS), a metric that quantifies layout diversity under fixed constraints. Moreover, to improve geometric consistency, we introduce a Boundary Cross-Attention (BCA) module that enables conditioning on building boundaries. Our experiments show that BCA significantly improves boundary adherence, while prolonged training drives diversity collapse undiagnosed by FID, revealing a critical trade-off between realism and diversity. Out-Of-Distribution evaluations further demonstrate the models' reliance on dataset priors, emphasizing the need for generative systems that explicitly balance fidelity, diversity, and generalization in architectural design tasks.