3D Expression Base Generation for Dynamic Facial Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing 3D human face reconstruction technologies produce static models with limited implementable functions, lacking the ability to generate diverse expressions.
Innovation Solution
A method to generate a set of expression bases from a reconstructed 3D human face model, using image pairs of a target object in various head postures, allowing for the creation of high-precision, drivable 3D human face models that can be rendered in any expression, enabling more diversified product functions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a static 3D human face model is reconstructed from 2D images, then the reconstruction process is simple and fast, but the model has limited implementable functions and cannot generate diverse expressions
Solution Approach 1:
The patent transforms the static 3D face model into a dynamic system by introducing expression bases that enable the model to generate diverse facial expressions. The 3D human face model is enhanced with expression capabilities through parameter-driven deformation, allowing it to transition between different expression states while maintaining the original reconstruction efficiency.
Solution Approach 2:
The patent employs parameter changes to control facial expressions by adjusting expression coefficients that modify the 3D face model geometry. By varying these parameters, the system can generate different expressions (e.g., smiling, frowning, surprised) from the same base model, thus achieving expression diversity without requiring multiple separate models.
2Adaptability or versatility
If expression bases are added to enable diverse expressions, then the model functionality is enhanced, but the system complexity increases
Solution Approach 1:
The patent segments the facial geometry into distinct expression bases corresponding to different facial muscle movements. Each expression base represents a specific deformation pattern (e.g., mouth opening for smiling, eyebrow raising for surprise). This segmentation allows independent control of different facial regions and expressions, managing complexity through modular organization.
Solution Approach 2:
The patent creates a universal 3D face model structure that can generate multiple expressions through a single set of expression bases. Rather than requiring separate models for each expression, the system uses one multi-functional model that can adapt to various expression requirements through parameter adjustment, thereby reducing overall system complexity.
3Manufacturing precision
If high-precision 3D human face models are generated with drivable expression bases, then the model accuracy and functionality are improved, but the processing time and computational resources increase
Solution Approach 1:
The patent performs preliminary action by pre-computing and storing expression bases during the model creation phase. These expression bases are prepared in advance and stored as part of the 3D face model data structure. When expression generation is needed, the system simply retrieves and applies the pre-computed bases with minimal additional processing, thus maintaining high precision while reducing real-time processing time.
Data Source
AI summary
This application provides a three-dimensional (3D) expression base generation method performed by a computer device. The method includes: obtaining image pairs of a target object in n types of head postures, each image pair including a color feature image and a depth image in a head posture; constructing a 3D human face model of the target object according to then image pairs; and generating a set of expression bases of the target object according to the 3D human face model of the target object. According to this application, based on a reconstructed 3D human face model, a set of expression bases of a target object is further generated, so that more diversified product functions may be expanded based on the set of expression bases.