3D Partial Model Completion via Symmetry Inference
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current 3D scanning technologies face challenges in generating complete point clouds, particularly with LiDAR devices, as they often capture only visible surfaces, leading to incomplete models and difficulties in calculating surface areas and volumes, which is time-consuming and impractical for applications like furniture arrangement and medical aid fitting.
Innovation Solution
A method and apparatus for computer-aided completion of 3D partial models by determining the geometry of the object through comparison with comparable objects or mirrored models, identifying hidden regions, and supplementing the model to achieve completeness, using 3D object recognition and neural networks for accuracy and efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If LiDAR devices are used for fast 3D scanning, then scanning speed is improved, but completeness of the point cloud deteriorates
Solution Approach 1:
The patent creates virtual copies of visible object surfaces by generating 3D models from the captured point cloud data. These virtual copies represent hidden surfaces that cannot be directly observed, effectively copying the geometry of occluded regions based on visible portions and symmetry relationships.
Solution Approach 2:
The patent detects and exploits symmetry relationships between different portions of objects. By identifying symmetric patterns in the visible point cloud data, the system can infer the geometry of hidden surfaces through symmetric transformation, converting the asymmetric limitation of single-view scanning into a solvable geometric problem.
2Measurement precision
If manual post-processing is performed to complete 3D models, then measurement accuracy is improved, but time consumption increases
Solution Approach 1:
The system performs automated completion of hidden surfaces using algorithmic processing of the captured point cloud data. The computer implementation automatically detects symmetry, generates virtual surfaces, and calculates measurements without requiring manual intervention, making the system self-sufficient in completing the 3D model reconstruction task.
Solution Approach 2:
The patent replaces manual mechanical post-processing operations with computer-based automated algorithms. The system uses computational methods to detect symmetry, generate virtual surfaces, and calculate measurements, substituting human manual work with automated image processing and geometric calculation algorithms.
3Loss of information
If complete scanning of all surfaces is attempted, then model completeness is improved, but ease of operation deteriorates
Solution Approach 1:
The patent performs scanning from a limited number of positions (partial action) rather than requiring complete 360-degree coverage. By capturing point clouds from a few strategic viewpoints and using symmetry-based inference, the system achieves complete model reconstruction without the excessive action of scanning every surface from every angle.
Solution Approach 2:
The system introduces virtual surfaces as an intermediary representation to bridge the gap between visible and hidden portions of objects. These computationally generated surfaces act as mediators that connect observed point cloud data with unobserved regions, enabling complete model reconstruction without direct physical access to all surfaces.
Data Source
AI summary
A method for the computer-aided completion of a 3D partial model—formed by points—of a partial region of an object that is captured by at least one capture device, wherein the 3D partial model can be supplemented with a hidden or missing partial region of the object situated outside the 3D partial model of the object that is to be completed, is provided. The method includes determining a geometry of the object, identifying the hidden or missing partial region of the object on the basis of the determined geometry of the object, supplementing the 3D partial model to form a complete 3D model with the identified hidden or missing partial region of the object, and c) outputting the completed 3D model at an output unit.