Scopus
A Framework for Architectural Geometry Synthesis Using Cellular Automata and Conditioned Diffusion Models
Proceedings SIGGRAPH Asia 2025 Posters SA 2025 · Aralık 2025
Makale Bilgileri
DergiProceedings SIGGRAPH Asia 2025 Posters SA 2025
Yayın TarihiAralık 2025
Scopus ID2-s2.0-105029855824
Özet
We present a framework for generating 3D architectural geometry and corresponding high-fidelity renderings from procedural rules. Our pipeline integrates a 3D Cellular Automata (CA) engine for voxel massing, a geometry-conditioned diffusion model (SDXL with ControlNet) for structure-aware rendering, and a Large Language Model (LLM) for automated metadata generation. This system links procedural logic with visual design, enabling rapid exploration and the creation of structured datasets. We demonstrate the framework by generating over 15,000 unique configurations at 70 seconds per model.
Yazarlar (3)
1
Sabri Gökmen
2
Andrei Vince
ORCID: 0009-0003-6422-4142
3
Aidan Oh
ORCID: 0009-0007-1789-8231
Anahtar Kelimeler
architecture
cellular automata
computational design
ControlNet
diffusion models
generative AI
procedural generation
prompt engineering
Kurumlar
The University of North Carolina at Chapel Hill
Chapel Hill United States
The University of North Carolina at Charlotte
Charlotte United States