Back to Wiki

[79] BRIDGING BIM AND AI: A Graph-BIM Encoding Approach For Detailed 3D Layout Generation Using Variational Graph Autoencoder

Seed Paper Structural Design & Layout Generation 2024

Paper Description extracted from PDF · abstract · 132 words

Building Information Modelling (BIM) data provides an abundant source with hierarchical and detailed info rmation on architectural elements. Nevertheless, transforming BIM data into an understandable format for AI to learn and generate controllable and detailed three-dimensional (3D) models remains a si gnificant research challenge. This paper explores an encoding approach for conver ting BIM data into graph-structured data for AI to learn 3D models, which we define as Graph-BIM encoding. We employ the graph reconstruction capabilities of a Variational Graph Autoencoder (VGAE) for the unsupervised learning of BIM data to identify a suitable encoding method. VGAE's graph generation capabilities also reason for spatial layouts. Results demonstrate that VGAE can reconstruct BIM 3D models with high accuracy, and can reason the en tire spatial layout from partial layout information detailed with architectural components.

AuthorsJiadong Liang and Ximing Zhong and Immanuel Koh
Year2024
VenueProceedings of the International Conference on Computer-Aided Architectural Design Research in Asia
DOI10.52842/CONF.CAADRIA.2024.1.221
Source DatabaseSeed (CumInCAD)
Bridge-to-GNNCategory A
GNN ArchitectureGraph GAN (heuristic — verify)
Graph EncodingBIM Element Topology Graph (inferred from title)
AEC TaskFloorplan / Layout Generation (inferred from title)
CohortMature Applications (2024–2026)

Extracted Methodology heuristic — verify before citing

Implementation FrameworkRevit / Dynamo (heuristic — verify)
Primary MetricAccuracy (heuristic — verify)
Key Findingincorporating room function connections as special attributes within our encoding framework (heuristic — verify)
Quality Assessment1/6 · rigor: Low ·
Baseline: ✓ (heuristic — verify)

Detected Technical Stack auto-detected from PDF · regex catalogue match

GNN architectures detectedGraph GAN · Graph VAE / VGAE · Knowledge Graph Embedding (TransE/RotatE/DistMult) · GCN
Frameworks / librariesRevit / Dynamo
Quality (heuristic, 6-flag)0/6 · rigor: Low
Code/Data: —Benchmark: —Baseline: —CV/Split: —Ablation: —Reproducible: —

Lineage Families

floorplan-generation

Enabled By / Precursors

[53] [88]

Extended By

[11]

Infrastructure Dependencies

BIM (auto-suggested) Revit / Dynamo (detected from PDF)

Research Frontier Tags

bim-graph

Part of the GML/GNN in AEC Systematic Review (PRISMA 2020) — 112 papers, 2020–2026