3D Mesh Texture Merging Using Confidence-Weighted Alignment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The process of creating 3D models from scans often results in the loss of color and texture information during alignment and merging, leading to issues like ghost images, fuzzy edges, and unclear models due to incorrect application of average color and texture.
Innovation Solution
A method that determines a confidence interval for texture information using vectors and weights the color values based on this confidence to accurately align and apply textures to the 3D mesh, reducing the likelihood of ghost images and improving model clarity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If average color and texture are applied to locations on the 3D mesh, then color and texture information is preserved, but ghost images and fuzzy edges occur due to incorrect application
Solution Approach 1:
The patent applies different weighting strategies to different locations on the 3D mesh based on local geometric properties and scan quality. Each vertex receives weighted contributions from multiple scans according to its specific characteristics, allowing accurate texture application in some regions while handling problematic areas differently to avoid ghost images and fuzzy edges.
Solution Approach 2:
The patent changes the parameter of texture application by introducing confidence intervals and weighting factors that vary from vertex to vertex. Instead of uniform averaging, the system dynamically adjusts the contribution of each scan based on local geometry, scan quality, and vertex visibility, thereby improving texture accuracy while preventing artifacts.
2Shape
If multiple scans are merged to create a 3D mesh, then complete geometry is achieved, but color and texture information is lost during alignment and merging
Solution Approach 1:
The patent performs preliminary actions by calculating confidence intervals and determining optimal texture sources before final texture application. The system pre-processes the scan data to identify which scans provide the most reliable texture information for each vertex, then uses this information to guide the texture mapping process, ensuring both geometric accuracy and texture preservation.
Solution Approach 2:
The patent implements feedback mechanisms by using confidence intervals to evaluate the quality of texture information from different scans. The system continuously assesses which scans provide reliable texture data and adjusts the weighting accordingly, creating a feedback loop that optimizes texture preservation while maintaining geometric accuracy throughout the merging process.
Data Source
AI summary
Examples disclose a method and system for merging textures. The method may be executable to receive one or more images of an object, identify a texture value for a point in a first image of the one or more images, and determine a metric indicative of a relation between a view reference point vector and a normal vector of a position of a point on the object relative to the image capturing device. Based on the metrics, the method may be executable to determine a weighted average texture value to apply to a corresponding point of a three-dimensional mesh of the object.