3D Shape Generation via Cross-Sectional Decomposition and Basic Shape Fitting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for generating three-dimensional computer graphics models in VR and AR require significant user input, imposing a heavy burden on users due to the manual specification of vertexes and sides for multiple images, which is time-consuming and labor-intensive.
Innovation Solution
An image processing apparatus that generates three-dimensional shape data by decomposing the schematic shape of an object into partial data based on cross-sectional views, fitting basic shapes such as cuboids, spheres, and cylinders, and integrating these shapes to produce accurate three-dimensional models with reduced user input.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual specification of vertexes and sides is used for multiple images, then three-dimensional CG model generation is achieved, but user burden increases and processing time extends
Solution Approach 1:
The patent segments the three-dimensional model generation process into multiple stages: first generating a rough three-dimensional shape from silhouettes, then decomposing it into cross-sectional data, and finally fitting basic shapes to each cross-section. This segmentation allows automated processing to handle the time-consuming aspects while maintaining accuracy requirements.
Solution Approach 2:
The patent introduces cross-sectional data as an intermediary representation between the rough three-dimensional shape and the final fitted model. By decomposing the complex three-dimensional shape into two-dimensional cross-sections, the system enables automated basic shape fitting while preserving the accuracy needed for precise three-dimensional reconstruction.
2Measurement precision
If manual specification of vertexes and sides is used for multiple images, then three-dimensional CG model generation is achieved, but user burden increases
Solution Approach 1:
The patent implements self-service by enabling the system to automatically generate three-dimensional models from captured images without requiring manual vertex and side specification. The automated pipeline includes silhouette extraction, three-dimensional shape generation, cross-sectional decomposition, and basic shape fitting, all performed without user intervention beyond initial image capture.
Solution Approach 2:
The patent replaces the manual mechanical process of specifying vertexes and sides with an automated computational system. Instead of users manually defining geometric features, the system uses image processing algorithms to extract silhouettes, generate three-dimensional shapes, and fit basic shapes automatically.
3Productivity
If basic shape fitting is performed on the entire schematic shape, then three-dimensional shape data is generated, but accuracy decreases for complex objects
Solution Approach 1:
The patent segments the object into multiple cross-sectional sections and applies basic shape fitting to each section independently rather than attempting to fit a single basic shape to the entire object. This segmentation enables accurate representation of complex objects with varying geometries along different axes.
Solution Approach 2:
The patent applies local quality by allowing different basic shapes to be fitted to different cross-sections based on their local characteristics. Each cross-section can have its own optimal basic shape fitting, enabling the system to capture local geometric variations while maintaining overall three-dimensional accuracy.
Data Source
AI summary
The image processing apparatus includes: a shape generation unit configured to generate data indicating a schematic shape of an object; a shape decomposition unit configured to decompose the data indicating the schematic shape of the object into a plurality of pieces of partial data in accordance with a shape of a cross section of the schematic shape of the object; and a shape fitting unit configured to fit a corresponding basic shape for each piece of the partial data, and generates three-dimensional shape data on the object based on the fitted basic shape.