Indoor Space Model Generation from Vectorized Drawings
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing metaverse platforms struggle to automatically generate accurate indoor space models replicating real-world environments, requiring significant time and expertise from professionals.
Innovation Solution
An indoor space model generation apparatus that derives information from vectorized drawing images, generates space configuration models in tile units, performs tile modeling, and creates indoor space models, enabling modular and automatic modeling of real indoor spaces.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If experts directly intervene in modeling operations to replicate real-world spaces, then modeling accuracy is improved, but time consumption and cost increase
Solution Approach 1:
The patent uses drawing images (blueprints, floor plans) as copies of real-world spaces to automatically generate 3D models. Instead of experts manually modeling from scratch, the system extracts spatial information from these 2D drawing copies and converts them into accurate 3D representations, significantly reducing time while maintaining precision through automated image processing and mesh generation algorithms
Solution Approach 2:
The patent replaces the manual mechanical process of expert modeling with an automated computational system. The system uses computer vision algorithms, image processing, and automated mesh generation to substitute human experts' manual operations, achieving both high accuracy and efficiency through automated spatial information extraction and 3D model construction
2Ease of manufacture
If conventional drawing-based modeling methods are used to generate mesh models, then modeling process is simplified, but editing flexibility is reduced
Solution Approach 1:
The patent segments the 3D model into modular components such as walls, floors, ceilings, and furniture elements that can be independently edited. Each component is represented as a separate mesh object with its own properties, allowing users to modify individual elements without affecting the entire model, thus maintaining both process simplicity and editing flexibility
Solution Approach 2:
The patent implements dynamic editing capabilities where the mesh model structure can be adaptively modified during the modeling process. Users can add, remove, or modify model elements dynamically, and the system automatically updates the overall model structure, providing flexibility while maintaining the simplicity of the drawing-based workflow
3Productivity
If automatic modeling is implemented without expert intervention, then productivity is improved, but modeling precision deteriorates
Solution Approach 1:
The patent incorporates feedback mechanisms where the automated system generates initial 3D models from drawing images, then uses validation algorithms to check geometric consistency, spatial relationships, and architectural合理性. The system iteratively refines the model by detecting and correcting errors, ensuring high accuracy while maintaining automated high-speed processing
Solution Approach 2:
The patent implements self-service capabilities where the automated modeling system independently performs spatial information extraction, mesh generation, and model validation without human intervention. The system uses built-in algorithms to automatically handle complex processing tasks, achieving both high productivity and precision through intelligent automated decision-making and error correction
Data Source
AI summary
an indoor space model generation apparatus for generating an indoor space model includes a derivation unit configured to derive, from a vectorized drawing image, information about space configuration objects included in the drawing image; a space configuration model generation unit configured to generate a space configuration model for each space configuration object in a tile unit based on the information about the space configuration objects; a tile modeling unit configured to perform tile modeling on the generated space configuration model; and an indoor space model generation unit configured to generate the indoor space model by using the space configuration model on which the tile modeling has been performed.