Automated 3D Modeling from 2D Floor Plans
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Creating 3D models from 2D floor plans is time-consuming and requires specialized knowledge and software, making it difficult for laypersons to convert 2D images into immersive 3D experiences, especially in real estate and other industries, where customization and cost-effectiveness are challenges.
Innovation Solution
A system and method utilizing machine learning models for symbol detection and segmentation, combined with 3D computer graphics software like Blender, to automatically generate 3D files from 2D architectural floor plans, allowing for the inclusion of structural features, furniture, and finishes, and exporting files in formats like .gltf, .obj, and .stl.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If traditional manual 3D modeling methods are used, then high-quality 3D models can be created, but the process requires years of education and months of work, making it inaccessible to laypersons
Solution Approach 1:
The patent replaces manual mechanical 3D modeling operations with an automated system that uses machine learning models (symbol detection model and segmentation model) to automatically generate 3D models from 2D floor plans. This substitution eliminates the need for specialized knowledge and extensive manual work, making the process accessible to laypersons while dramatically reducing creation time
Solution Approach 2:
The system enables self-service 3D modeling by allowing users to upload their own 2D floor plan images and automatically receive generated 3D models without requiring professional intervention. The automated processing pipeline handles symbol detection, segmentation, and model generation independently, empowering users to create their own 3D models without hiring specialists
2Manufacturing precision
If specialized software is used for 3D modeling, then accurate 3D models can be generated, but the software is prohibitively expensive and requires specialized knowledge
Solution Approach 1:
The patent replaces complex specialized 3D modeling software with an automated machine learning-based system. Instead of requiring users to manually operate complex software tools, the system uses trained models to automatically detect symbols, segment structures, and generate 3D models, maintaining accuracy while eliminating software complexity and cost barriers
Solution Approach 2:
The system creates 3D models by copying and extruding 2D floor plan structures into three-dimensional space. The symbol detection model identifies 2D symbols representing building elements, and the system automatically copies these into 3D space with appropriate geometric properties, achieving accurate 3D reconstruction without requiring complex modeling software
3Adaptability or versatility
If virtual tours are created from panoramic photographs, then 3D visualization is achieved, but the space cannot be easily customized with furniture or alternate nonstructural component locations
Solution Approach 1:
The patent applies segmentation by dividing the 3D model into distinct structural and nonstructural components. The segmentation model identifies and separates walls, windows, doors, and other elements as independent objects, allowing them to be individually modified, moved, or replaced without affecting the entire model, thus enabling easy customization while maintaining manageable complexity
Solution Approach 2:
The system creates dynamic 3D models where nonstructural components such as furniture, interior walls, and fixtures can be easily repositioned and reconfigured. Unlike static virtual tours, the generated models allow users to modify element locations and configurations, providing adaptability while the automated generation process keeps the underlying system relatively simple
4Productivity
If manual 3D modeling is performed for multiple spaces, then customized 3D models can be created, but it requires hiring specialized personnel for each space, increasing time and cost
Solution Approach 1:
The patent creates a universal 3D modeling system that can process multiple different floor plans through the same automated pipeline. The symbol detection and segmentation models are designed to handle various building types and layouts, allowing rapid generation of 3D models for multiple spaces without requiring specialized personnel for each project, thereby dramatically increasing productivity while reducing time investment
Data Source
AI summary
Disclosed are a system and method for generating and viewing a 3D file from a 2D architectural floorplan. The method includes the steps of: uploading a 2D file of an architectural floorplan; inputting default structural settings corresponding to structural features of the architectural floorplan; executing a symbol detection model on the 2D file so as to detect symbols present in architectural floorplan; executing a segmentation model on the 2D file so as to identify segments corresponding to walls and windows; vectorizing the identified segments resulting from the execution of the segmentation model; and generating a 3D file based on the results of the symbol detection model and the segmentation model. The resulting 3D file is viewable in a 3D computer graphics engine. The system includes a frontend adapted to receive the 2D architectural floorplan file, and a backend adapted to enqueue worker instances including the segmentation and symbol detection models.