GFLAN: Generative Functional Layouts

2025-12-18

AI Summary

To address the challenge of jointly modeling topological reasoning, geometric constraints, and functional requirements in architectural floor plan generation, this paper proposes a decoupled two-stage framework: topological planning followed by geometric realization. First, room centroids are probabilistically sampled within the building outline to encode topological relationships. Second, a room-boundary heterogeneous graph is constructed, and a dual-encoder CNN—processing spatial invariance and layout dynamics separately—is employed; a Transformer-enhanced GNN then jointly regresses room boundaries, enabling functional adjacency propagation and emergent circulation patterns. This work is the first to explicitly separate topological modeling from geometric instantiation, significantly improving performance on multi-scale residential layouts in terms of functional validity and topological connectivity. The method supports editable vector layout generation from minimal inputs—namely, a single boundary contour and an entry point—while ensuring regulatory compliance.

Technology Category

• Planning, Routing, and Scheduling: Planning with Language Models
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Cognitive Modeling & Cognitive Systems: Agent Architectures

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems

Abstract

Automated floor plan generation lies at the intersection of combinatorial search, geometric constraint satisfaction, and functional design requirements -- a confluence that has historically resisted a unified computational treatment. While recent deep learning approaches have improved the state of the art, they often struggle to capture architectural reasoning: the precedence of topological relationships over geometric instantiation, the propagation of functional constraints through adjacency networks, and the emergence of circulation patterns from local connectivity decisions. To address these fundamental challenges, this paper introduces GFLAN, a generative framework that restructures floor plan synthesis through explicit factorization into topological planning and geometric realization. Given a single exterior boundary and a front-door location, our approach departs from direct pixel-to-pixel or wall-tracing generation in favor of a principled two-stage decomposition. Stage A employs a specialized convolutional architecture with dual encoders -- separating invariant spatial context from evolving layout state -- to sequentially allocate room centroids within the building envelope via discrete probability maps over feasible placements. Stage B constructs a heterogeneous graph linking room nodes to boundary vertices, then applies a Transformer-augmented graph neural network (GNN) that jointly regresses room boundaries.

Problem

• Generates floor plans via topological and geometric factorization
• Addresses architectural reasoning in automated layout generation
• Sequentially allocates rooms and regresses boundaries using neural networks

Innovation

• Two-stage decomposition: topological planning then geometric realization
• Dual encoder CNN for sequential room centroid allocation
• Transformer-augmented GNN for joint room boundary regression