Constructive Intelligence: AI Floorplans 2D - 3D Current State of Tech

The transformative capabilities of artificial intelligence (AI) in architecture and real estate are gaining momentum, specifically through the use of machine learning (ML) and large language models (LLMs) to convert 2D floorplans into dynamic 3D models. The convergence of these technologies enables more efficient and engaging design processes and property visualisations. It is important to have 3D representations because they help bridge the imagination gap.

Having spent many hours as an architecture student myself, painstakingly creating 2D plans to have then start another process to create 3D (while this is improved with tools like Revit at the time of design), If you want to go from 2D schematic designs, the typical floor plan designs you might see on a property listing, to 3D there is often a lot of manual work involved. There are also two different fidelities: photorealism, which is used for marketing and can take time to render, and 3D for experimental representation, which often requires expertise in additional tools.

In this article, I will shed some light on how AI, both traditional Machine Learning and Multi-Model (can ingest images and text) Foundation Models, can save time and money and inject some joy into creating these important assets.

Please share tools you have seen and work with.

How AI Transforms 2D Floorplans to 3D Models

Machine Learning (ML) Techniques

The use of machine learning for converting 2D floorplans into 3D models is grounded in specialized algorithms that have been trained to understand spatial relationships, architectural semantics, and design patterns. Here’s how ML functions in this context:

1. Image Recognition and Segmentation: ML models, primarily using Convolutional Neural Networks (CNNs), detect walls, doors, windows, and other key features within a floorplan. The CNN learns from a labeled dataset of floorplans and identifies structural elements accurately.
2. Spatial Reasoning: Using Graph Neural Networks (GNNs) or other spatial reasoning algorithms, the ML model understands the connectivity and relationship between different components of the plan, like room adjacencies or wall intersections.
3. Depth and Extrusion Calculation: ML models estimate the height of walls and dimensions of spaces. Techniques like monocular depth estimation allow the ML system to predict depth from a single flat image, making the extrusions necessary to convert a 2D outline into a 3D structure.

Example in Use: Autodesk uses these ML principles in tools like Revit, which helps architects automate tedious tasks like generating structural components from a 2D schematic, saving significant time in the design process.

Large Language Models (LLMs) in Architectural Design

While LLMs like OpenAI’s GPT or Google Gemini models are not traditionally associated with spatial modelling, they are finding unique applications in the design and construction industry. Here’s how LLMs are making an impact:

1. Contextual Understanding and Guidance: LLMs can interpret complex architectural terminology and provide contextual recommendations. For instance, when fed a description of a floorplan along with functional requirements (like room purposes or accessibility needs), LLMs can assist in outlining an initial design concept or layout. Recommend you check out High Arc's demos here. Recognising a room and adding assets to it based on room classification was mentioned in their latest Webinar.
2. Text-to-Design Automation: One promising application is in automating the drafting process based on natural language descriptions. Imagine an architect describing a vision—“a three-bedroom house with a spacious living room, open kitchen, and a southern-facing balcony”—and an LLM-guided system generating a preliminary floorplan that can then be refined into a 3D model.
3. Code and Compliance Assistance: LLMs are adept at parsing and understanding text, making them ideal for ensuring that a design adheres to local zoning and building regulations. Integrating LLMs into architectural software allows for real-time feedback on compliance.

Example in Use: Higharc is exploring how LLMs can work in tandem with ML models to automate design compliance checks and generate design variations based on user input, which is beneficial in the early planning stages. (Beta-stages)

Comparison: ML vs. LLMs in 2D to 3D Conversion

Integrating ML and LLMs for Enhanced Design Processes

Combining ML and LLM capabilities holds immense promise for the architecture and real estate sectors:

1. Enhanced User Experience: By integrating LLMs into an ML-driven design platform, architects and users can interact with the system through natural language. For instance, an architect could verbally specify modifications to a design, and the ML algorithms would adjust the 3D model accordingly.
2. Automated Documentation: LLMs can draft detailed architectural reports and specifications, reducing the time architects spend on paperwork. ML algorithms, meanwhile, continue refining the 3D model for accuracy and visualization.
3. Smart Compliance and Suggestions: ML ensures the structural and spatial correctness of models, while LLMs offer insights into potential design improvements and adherence to regulations.

Example Collaboration: PLACE Technologies is using LLMs in their Digital Housing Developer Archer to not only generate 3D models from 2D but also provide text-based interaction to allow the purchaser of an MMC development to customise the model with text commands.

Benefits of AI-Driven 3D Modeling

1. Time and Cost Savings: Automating the conversion of 2D plans to 3D models allows architects to focus more on creative and strategic aspects of their work. Property developers can also reduce costs by minimizing the need for manual adjustments and reworks.
2. Better Client Engagement: Real estate marketers benefit from immersive 3D visualizations, which make properties more appealing and understandable to potential buyers/renters.
3. Increased Accuracy and Efficiency: ML reduces human error in translating floorplans, while LLMs facilitate communication and compliance checking, creating a holistic design and marketing solution.

References and Further Reading

1. Higharc’s Generative Design Platform - Explore how Higharc uses AI to revolutionize home design with automated compliance and customizable floorplans.
2. PLACE Technologies - Discover PLACE’s approach to integrating AI into their global MMC Marketplace
3. Autodesk Revit - Learn how Autodesk leverages ML in its design tools to automate structural and material modeling.