AI in Your Toolbox: A Plugin for Generating Renderings from 3D Models

TL;DR

This work presents a Rhino platform plugin that harnesses stable diffusion to generate and edit architectural renderings directly from 3D models, enabling real-time AI deployment within CAD. It integrates Text-to-Image generation with ControlNet edge/depth constraints and LoRA-based style control to ensure stylistic consistency, while offering adjustable diffusion parameters and local editing capabilities for text-guided refinements. The approach supports cross-platform interoperability and rapid design exploration by enabling prompt-driven generation and editing from the Rhino environment. The contributions demonstrate how AI-enabled rendering can enhance architectural design and engineering planning in AI-driven CAD workflows, with practical implications for design efficiency and quality.

Abstract

With the rapid development of LLMs and AIGC technology, we present a Rhino platform plugin utilizing stable diffusion technology. This plugin enables real-time application deployment from 3D modeling software, integrating stable diffusion models with Rhino's features. It offers intelligent design functions, real-time feedback, and cross-platform linkage, enhancing design efficiency and quality. Our ongoing efforts focus on optimizing the plugin to further advance AI applications in CAD, empowering designers with smarter and more efficient design tools. Our goal is to provide designers with enhanced capabilities for creating exceptional designs in an increasingly AI-driven CAD environment.