3D Facial Morphing Using Learned Generic Head Model
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing facial morphing methods require high computer processing power and are typically limited to cosmetic treatment settings, making it difficult for patients to preview simulation outcomes before visiting a doctor.
Innovation Solution
A method for rendering a 3D digital image using a combination of algorithms such as ICP, TSDF, marching cubes, Poisson reconstruction, and texture mapping, along with a learned generic head model, to create a 3D mesh and stitch images with consistent lighting, enabling automated facial morphing on mobile devices without professional assistance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional 3D model morphing methods are used, then high-quality facial simulation is achieved, but high computer processing power and specialized algorithms are required, limiting availability to cosmetic treatment offices
Solution Approach 1:
The patent uses a pre-built generic 3D head model as a template that can be copied and adapted for different patients. Instead of creating a completely new 3D model for each patient, the system starts with a standardized model and modifies it to match patient-specific features, significantly reducing computational requirements while maintaining simulation quality
Solution Approach 2:
The system transforms the complex 3D modeling problem into a parameter optimization problem. By adjusting a limited set of morphable model parameters (such as shape coefficients and texture parameters) to match patient photographs, the system achieves accurate facial simulation without requiring full 3D scanning and modeling, thus reducing device complexity
2Measurement precision
If traditional facial morphing systems are used, then accurate facial simulation is achieved, but the systems are only available at cosmetic treatment offices, reducing patient accessibility
Solution Approach 1:
The patent enables patients to perform facial morphing simulations themselves using their own smartphones or computers. The system is designed to be self-service oriented, where patients can upload their photographs and generate simulations without requiring professional equipment or assistance, thereby dramatically improving accessibility
Solution Approach 2:
The system uses a universal generic head model that can be applied to patients of different ages, genders, and ethnicities. This multi-functional approach allows the same software system to serve diverse patient populations across different locations, eliminating the need for specialized office-based systems and enabling widespread accessibility
Data Source
AI summary
In one aspect, a computerized method for rendering a three-dimensional (3D) digital image for automated facial morphing includes the step scanning of the user's face with a digital camera to obtain a set of digital images of the user's face. The method includes the step of determining that a user's face is in a compliant state. The method includes the step of implementing an analysis of the set of digital images and implementing a set of pre-rendering steps. Each digital image comprises a depth data, a red/green/blue (RGB) data, and a facemask data. The method then implements an iterative closest path (ICP) algorithm that correlates the set of digital images together by stitching together the cloud of points of the facemask data of each digital image and outputs a set of transformation matrices. The method includes the step of implementing a truncated signed distance function (TSDF) algorithm on the set of transformation matrices. The TSDF algorithm represents each point of the transformation matrices in a regularized voxel grid and outputs a set of voxel representations as a one-dimension (1-D) array of voxels. The method includes the step of implementing a marching cubes algorithm that obtains each voxel representation of the 1-D array of voxels and creates a three-dimensional (3D) mesh out of the per-voxel values provided by the TSDF and outputs a mesh representation. The mesh representation comprises a set of triangles and vertices. The method comprises the step of implementing a cleaning algorithm that obtains the mesh representation and cleans the floating vertices and triangles and outputs a mesh. The mesh comprises a set of scattered points with a normal per point. The method includes the step of implementing a Poisson algorithm on the mesh output and fills in any holes of the mesh. The Poisson algorithm outputs a reconstructed mesh. The method fits the reconstructed mesh on a trained three-dimensional (3D) face model and a specified machine learning algorithm is used to fit the trained 3D face model to the 3D landmarks in the reconstructed mesh.