Virtual Head Modeling for Accurate Glasses Contact Fit
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing virtual modeling methods for user's heads require high computational effort and may not achieve sufficient accuracy for precise virtual placement of accessories like glasses, especially in contact areas, due to uniform morphing algorithms and insufficient precision.
Innovation Solution
A method involving a machine learning algorithm to identify specific head characteristics, combined with a rigid sequential iterative closest point algorithm and a cost function minimization process, to align and deform a statistical head model accurately, focusing on regions relevant for accessory placement while reducing computational cost.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If uniform morphing algorithms are used to deform the head model, then the modeling process is simple, but the accuracy in contact areas for accessory placement is insufficient
Solution Approach 1:
The patent segments the head model into multiple regions (contact areas and non-contact areas) and applies different morphing strategies to each. Contact areas use a controlled morphing algorithm that preserves geometric fidelity, while non-contact areas use standard morphing, thereby achieving both accuracy where needed and computational efficiency elsewhere.
Solution Approach 2:
The patent implements local quality by applying different morphing algorithms to different regions of the head model. Specifically, contact areas (where accessories like glasses will be placed) use a precision-preserving morphing algorithm, while other areas use standard morphing, ensuring high accuracy in critical regions without compromising overall process simplicity.
2Manufacturing precision
If high computational effort is expended to achieve precise virtual placement of accessories, then the accuracy improves, but the processing time and resources increase
Solution Approach 1:
The patent divides the head model into contact and non-contact regions, applying computationally intensive precision-preserving morphing only to contact areas where accessory placement accuracy is critical. This segmentation allows the system to achieve high precision where needed while maintaining fast processing times for the overall model.
Solution Approach 2:
The patent applies high-quality, computationally demanding morphing algorithms locally only to contact areas (such as the bridge of the nose and temples where glasses contact the head), while using simpler, faster algorithms for non-contact areas. This approach optimizes the balance between processing time and placement precision.
3Productivity
If the statistical head model is deformed using traditional algorithms, then the overall model is updated, but the regions relevant for accessory interaction do not achieve sufficient accuracy
Solution Approach 1:
The patent segments the head model deformation process into two distinct phases: first, a global deformation phase that updates the entire statistical head model efficiently; second, a localized refinement phase that applies precision-preserving morphing specifically to contact areas. This segmentation enables both efficient overall updating and high accuracy in critical regions.
Solution Approach 2:
The patent implements local quality by applying different deformation strategies to different regions. Contact areas (where accessories interact with the head) undergo controlled, precision-preserving deformation, while non-contact areas use standard efficient deformation algorithms, thereby achieving both productivity and precision in their respective domains.
Data Source
AI summary
A modelling method for making a virtual model of a user's head includes operations of: acquiring a plurality of depth images of a user's head in different head poses; processing the acquired depth images; localising on the two-dimensional images first marker points corresponding to characteristics of the head; estimating the head by estimating the yaw, pitch and roll angles of the head and the positioning of the head along three axes of a reference system of Cartesian axes for each of the two-dimensional images of the depth images, performing a transformation of the head pose; projecting the first marker points located in the respective cloud of points; associating iteratively in pairs the clouds of points until obtaining a cloud of synthesis points; aligning the cloud of synthesis points; and deforming the statistical head model by means of a minimization algorithm of a predefined cost function.