Temporary Texture Blending for Textureless 3D Reconstruction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing 3D modeling techniques struggle with textureless regions, leading to distorted or incomplete reconstructions due to the difficulty in tracking key points across different viewing angles.
Innovation Solution
Applying temporary textures to textureless regions using a neural radiance field model to enhance key point detection, followed by blending these textures with original images to generate a more accurate 3D surface model and excluding them from the final texture map.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If temporary textures are applied to textureless regions, then key point tracking accuracy is improved, but device complexity increases
Solution Approach 1:
The patent applies temporary textures to textureless regions before performing 3D reconstruction. This preliminary action adds artificial visual features to regions that would otherwise lack distinguishable characteristics, enabling key point detection algorithms to track points across multiple views accurately. The temporary textures are applied in advance to the input images, allowing the reconstruction process to proceed with improved feature matching.
Solution Approach 2:
The temporary textures serve as an intermediary element between the textureless regions and the key point detection algorithm. By introducing these artificial textures, the patent creates a mediator that bridges the gap between regions lacking natural features and the requirements of feature-based tracking, enabling accurate 3D reconstruction without modifying the underlying geometry or requiring additional hardware.
2Manufacturing precision
If temporary textures are blended with original images, then 3D surface model accuracy is improved, but processing time increases
Solution Approach 1:
The patent applies temporary textures selectively only to textureless regions rather than uniformly across the entire image. This local approach ensures that regions with sufficient natural texture remain unchanged, while only the problematic textureless areas receive artificial textures. The blending operation combines the temporary textures with original images in a localized manner, improving 3D surface model accuracy in specific regions without unnecessarily processing the entire image set.
Data Source
AI summary
Systems and methods are directed to generating a three-dimensional (3D) model of a first image set. The first image set (e.g., input image set) corresponds to different viewing angles of an object that is subject to 3D modeling. A volumetric density function is generated from the first image set. A second image set (e.g., a textured image set) is generated from the volumetric density function and from a predefined color function. The first image set is blended with the second image set to generate a third image set (e.g., an image set with temporary textures). To generate the 3D model, a 3D surface model is generated from the third image set. In addition, a texture map of the 3D surface model is generated from the first image set. A computing system is configured to render the 3D surface model and texture map for display.