Invert Spaces: Automated Building Element Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for automatically modeling building elements from scanned room data require manual adaptations and are prone to inaccuracies due to non-parallel room surfaces and small inclusions, leading to unreliable generation and classification of 3D architectural models.
Innovation Solution
A computer-implemented method that grows piecewise planar objects by extruding their planes with a user-defined thickness, forms a fused object, intersects it with additional planes to create a cell complex, classifies cells based on geometric criteria, and merges them into building elements such as exterior and interior walls, floors, and roofs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If automated segmentation methods (region growing, plane fitting, clustering) are used to divide point cloud data into building elements, then productivity is improved, but manufacturing precision deteriorates due to errors from non-parallel surfaces and small inclusions
Solution Approach 1:
The patent applies preliminary action by performing a first segmentation to identify candidate building elements, then using a second refinement segmentation to correct errors. This two-stage approach allows the system to quickly identify potential elements first, then refine them to achieve high precision, resolving the contradiction between automated speed and accuracy.
Solution Approach 2:
The patent implements feedback through an iterative refinement process where the output of the first segmentation becomes the input for the second segmentation. The system uses the initial segmentation results to guide subsequent refinement, continuously improving accuracy while maintaining automated operation. This feedback loop enables both high productivity and manufacturing precision.
2Manufacturing precision
If manual adaptations are applied to segmentation output to correct errors, then manufacturing precision is improved, but loss of time increases due to required manual intervention
Solution Approach 1:
The patent applies self-service by implementing an automated refinement segmentation that corrects errors without requiring manual intervention. The system uses geometric constraints and spatial relationships to automatically identify and correct segmentation errors, eliminating the need for time-consuming manual adjustments while maintaining high precision.
Solution Approach 2:
The patent replaces manual mechanical adjustment with an automated computational system. The refinement segmentation uses algorithmic processing to automatically correct segmentation errors, substituting human manual work with automated computational methods that are both faster and more consistent, thereby reducing time loss while improving precision.
3Manufacturing precision
If multiple segmentation techniques are combined in a hybrid approach, then manufacturing precision is improved, but device complexity increases
Solution Approach 1:
The patent applies segmentation by dividing the complex hybrid segmentation process into distinct, manageable stages: a first segmentation stage using multiple techniques to identify candidate elements, and a second refinement stage to correct errors. This segmentation of the processing workflow reduces overall system complexity while maintaining the precision benefits of combining multiple techniques.
Solution Approach 2:
The patent resolves complexity by transitioning from a single-complex-segmentation approach to a multi-stage hierarchical approach. The first segmentation operates at a coarser level to identify potential elements, while the second refinement operates at a finer level to correct specific errors. This dimensional hierarchy simplifies the overall process complexity while maintaining high manufacturing precision.
Data Source
Figure 1
Figure 2a
Figure 2b
AI summary
A computer implemented method for generating building elements from a set of piecewise planar objects that are obtained from scanning data from room interiors within a building. This method allows for the automatic generation and classification of building elements from a set of room interior representations, enabling an efficient modelling process in architectural and construction projects.