Virtual Object Generation Using Point Cloud Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for generating virtual objects from real objects in VR and AR content often fail to accurately represent the object's appearance, leading to errors in point cloud acquisition and conversion to 3D mesh, resulting in dissimilar virtual objects.
Innovation Solution
A method and apparatus that acquire a point cloud based on a depth map, determine shape attribute information using a neural network, calculate an energy field, and change point positions in the point cloud to generate a virtual object that accurately represents the real object's shape and appearance, incorporating texture generation from color images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional methods are used to generate virtual objects from real objects, then the generation process is simple, but the accuracy and similarity of the virtual object to the real object deteriorates
Solution Approach 1:
The generation process is divided into distinct segments: point cloud acquisition from depth maps, shape attribute determination using neural networks, energy field calculation, and point position optimization. This segmentation allows each stage to be optimized independently, improving overall accuracy while managing complexity through modular processing
Solution Approach 2:
Shape attribute information is determined in advance using neural networks before the actual point cloud optimization. This preliminary action provides guiding information for subsequent energy field calculations and point position adjustments, enabling more accurate virtual object generation without requiring complex real-time processing
2Reliability
If point cloud acquisition and conversion methods are simplified, then the processing speed is faster, but the similarity between virtual and real objects deteriorates
Solution Approach 1:
The patent transforms the point cloud representation by changing point positions based on energy field calculations derived from shape attribute information. This parameter transformation approach maintains geometric fidelity while enabling efficient processing through mathematical optimization rather than complex geometric operations
Solution Approach 2:
Traditional geometric processing methods are replaced with physics-inspired energy field calculations. The optimization of point positions is achieved through energy minimization principles rather than mechanical or iterative geometric adjustments, providing both accuracy and computational efficiency
Data Source
AI summary
A method and apparatus for generating a virtual object are provided, the method includes acquiring a point cloud of an object to generate a virtual object, determining shape attribute information of the object based on an image of the object, changing a position of at least one point in the point cloud based on the shape attribute information, and generating a virtual object for the object based on a changed point cloud.