3D Body Scan Imaging Without Texture Mapping Artifacts
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing three-dimensional scan systems face challenges in achieving high-resolution, natural representations of body parts, particularly skin and hair, due to texture mapping artifacts and resource-intensive modeling, which compromises monitoring and cosmetic treatment accuracy.
Innovation Solution
A system that obtains images and depth maps from multiple viewing directions, constructs a three-dimensional model, and presents augmented images with user input to enhance visualization, without texture mapping, maintaining high image quality and allowing 3D perception through sequential image display.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If texture mapping is used to create high-resolution 3D models, then image resolution is improved, but artifacts and unnatural appearance are introduced
Solution Approach 1:
The patent extracts and removes the texture mapping process from the 3D visualization system. Instead of mapping 2D textures onto 3D polygon models, the system directly processes and displays high-resolution 2D images, eliminating the source of texture mapping artifacts while preserving image quality and natural appearance.
Solution Approach 2:
The patent inverts the conventional approach by not creating a 3D model from 2D images for display purposes. Instead, it uses the 2D images themselves as the primary visualization medium, leveraging their inherent high resolution and natural appearance without the need for problematic texture mapping operations.
2Measurement precision
If high-resolution 3D models with many polygons are created, then image resolution is improved, but computational resources and processing time are excessively consumed
Solution Approach 1:
The patent uses lightweight 3D models as projection templates rather than creating high-resolution polygon models. These simplified 3D copies serve only to define projection geometry and map 2D images to 3D viewing directions, consuming minimal computational resources while achieving the desired 3D visualization effect.
Solution Approach 2:
The patent replaces the computationally intensive mechanical process of rendering high-resolution 3D polygon models with textures with a more efficient system that projects 2D images onto lightweight 3D structures. This substitution dramatically reduces processing time and computational resource requirements.
3Measurement precision
If 3D modeling is performed at very high resolution, then inspection quality is improved, but the natural appearance of skin and hair is lost
Solution Approach 1:
The patent applies different quality characteristics to different aspects of the system: high resolution is maintained in the 2D image domain for inspection quality, while the 3D model uses low resolution only for geometric projection. This local differentiation allows each component to optimize its own quality metrics without compromising the other.
Solution Approach 2:
The patent resolves the resolution conflict by separating the 2D image dimension (where high resolution is maintained for natural appearance) from the 3D model dimension (where low resolution is acceptable for structural purposes). This dimensional separation allows high-resolution images to be displayed naturally while the 3D model serves only as a projection framework.
4Adaptability or versatility
If texture mapping with scaling and deforming is applied, then 3D visualization is achieved, but monitoring accuracy of skin structures is compromised
Solution Approach 1:
The patent removes the texture mapping operations (scaling, deforming, patching) from the system. Instead of transforming 2D images to fit 3D model surfaces, the system projects images directly according to camera viewing directions, preserving the original image geometry and ensuring monitoring accuracy is not compromised.
Data Source
AI summary
A system (SYS) for obtaining a three-dimensional (3D) scan of a body part or a complete body to enable inspection of at least a part of the body part or the complete body is described. The system has an image acquisition system (IAQ) configured for obtaining a plurality of images (IMGS) from the body part or the complete body from a first plurality of first viewing directions and configured for obtaining a plurality of depth maps (DIMGS) from the body part or the complete body from a second plurality of second viewing directions. The system has an image processor (IPROC) configured for determining a three-dimensional model of the body part. The system has an augmenting unit (AUU) configured for using the three-dimensional model and augmented visual information to augment at least part of the plurality of images with the augmented visual information to obtain a plurality of augmented images, and configured for presenting a sequence of at least part of the plurality of augmented images on the display unit allowing to perceive a three-dimensional visualization of the body part or the complete body as augmented with the augmented visual information.