Animation Libraries for Facial Expression Identification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current computer-based animation techniques require numerous facial markers and optimal lighting conditions to capture detailed facial expressions, which are time-consuming and impractical for capturing motion in various environments, and often result in low-resolution motion meshes that do not accurately represent the actor's face.
Innovation Solution
A system that uses a library of previously captured animation information to identify and transfer facial expressions by comparing surface features to a motion model, allowing for a high-resolution animation mesh to be created with less data, using principal component analysis to decompose and compress motion information for efficient storage and reuse.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If numerous facial markers are applied to capture detailed facial expressions, then measurement precision is improved, but device complexity and time consumption increase
Solution Approach 1:
The patent uses a library of pre-computed motion models as copies of facial motion patterns. Instead of capturing every facial movement with numerous markers, the system compares captured surface features against these pre-stored motion models to identify and transfer facial expressions, reducing the need for extensive marker application while maintaining expression accuracy
Solution Approach 2:
The system performs preliminary action by pre-computing and storing motion models from captured images before actual use. These preprocessed motion models are stored in a library and can be quickly retrieved and applied to new captures, eliminating the need to reprocess extensive marker data for each new facial expression capture
2Measurement precision
If numerous facial markers are applied to capture detailed facial expressions, then measurement precision is improved, but loss of time increases
Solution Approach 1:
The patent uses a library of pre-computed motion models as copies of facial motion patterns. Instead of capturing every facial movement with numerous markers, the system compares captured surface features against these pre-stored motion models to identify and transfer facial expressions, reducing the need for extensive marker application while maintaining expression accuracy
Solution Approach 2:
The system performs preliminary action by pre-computing and storing motion models from captured images before actual use. These preprocessed motion models are stored in a library and can be quickly retrieved and applied to new captures, eliminating the need to reprocess extensive marker data for each new facial expression capture
3Device complexity
If a reduced set of surface features is used for identification, then device complexity is reduced, but measurement precision deteriorates
Solution Approach 1:
The patent uses a library of pre-computed motion models as copies of facial motion patterns. Instead of capturing every facial movement with numerous markers, the system compares captured surface features against these pre-stored motion models to identify and transfer facial expressions, reducing the need for extensive marker application while maintaining expression accuracy
Solution Approach 2:
The system changes the parameter approach by transforming raw surface feature data into decomposed principal components. This parameter transformation allows the system to work with reduced feature sets while maintaining identification accuracy, as the principal components capture the essential motion patterns in a compressed representation
4Manufacturing precision
If motion data is stored in high resolution for animation, then manufacturing precision is improved, but loss of information increases
Solution Approach 1:
The system changes the parameter approach by transforming raw surface feature data into decomposed principal components. This parameter transformation allows the system to work with reduced feature sets while maintaining identification accuracy, as the principal components capture the essential motion patterns in a compressed representation
Solution Approach 2:
The system extracts only the essential motion information from the full motion data by using principal component analysis. The motion models store only the critical deformation patterns and motion trajectories needed for animation, discarding redundant information while preserving the essential facial expression characteristics for high-resolution animation
Data Source
AI summary
A computer-implemented method includes comparing one or more surface features to a motion model. The surface feature or surface features represent a portion of an object in an image. The method also includes identifying a representation of the object from the motion model, based upon the comparison.