FACS Cleaning for Motion Capture Marker Noise
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Facial motion capture systems face challenges in accurately tracking and stabilizing facial marker data due to the subtlety of human facial expressions and the proximity of markers, leading to tracking errors, noise, and manual processing requirements.
Innovation Solution
The implementation of a Facial Action Coding System (FACS) that decimates frames into key facial poses, generates a facial pose matrix for constraint-based cleaning, and applies filtering to ensure accurate and smooth tracking of facial marker data points, constraining movements within acceptable limits and eliminating noise artifacts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a large number of markers are used to capture subtle facial expressions, then measurement precision is improved, but device complexity and difficulty of tracking increase
Solution Approach 1:
The patent segments facial markers into multiple groups (e.g., upper face, lower face, jaw) and processes each group separately through automated algorithms. This segmentation reduces the computational complexity of tracking all markers simultaneously while maintaining precision in capturing subtle facial expressions by focusing processing power on smaller marker subsets.
Solution Approach 2:
The patent introduces automated cleaning algorithms as intermediary processing steps between raw marker capture and final expression analysis. These algorithms act as mediators that automatically detect and correct tracking errors, reducing the need for manual intervention while preserving measurement precision from the large number of markers.
2Measurement precision
If manual processing is applied to ensure accurate marker tracking, then measurement precision is improved, but productivity and time consumption decrease
Solution Approach 1:
The patent implements self-service automated algorithms that independently detect, identify, and correct tracking errors in facial marker data without requiring manual intervention. The system uses cleaning algorithms that automatically process marker trajectories, eliminating the need for time-consuming manual review while maintaining high tracking accuracy through algorithmic error detection and correction.
Solution Approach 2:
The patent replaces manual mechanical processing with automated computational algorithms. Instead of human operators manually reviewing and correcting marker tracking data, the system employs computer-based cleaning algorithms that automatically process the data, significantly increasing productivity while maintaining or improving measurement precision through consistent algorithmic application.
3Measurement precision
If frame-by-frame marker tracking is performed, then measurement precision is improved, but loss of time increases
Solution Approach 1:
The patent applies preliminary cleaning and stabilization algorithms to marker data before detailed frame-by-frame analysis. By pre-processing the data to correct obvious tracking errors and stabilize marker positions in advance, the system reduces the time required for subsequent detailed analysis while maintaining measurement precision. This preliminary action prevents time-wasting re-processing of already-corrected data.
Solution Approach 2:
The patent implements continuous automated cleaning and stabilization processes that operate throughout the entire frame sequence rather than requiring discrete manual intervention at each frame. This continuous automated processing maintains measurement precision across all frames while significantly reducing total processing time compared to traditional manual frame-by-frame review methods.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method of cleaning facial marker data, the method comprising: decimating frames of survey facial marker data into representative frames of key facial poses; generating a facial pose matrix using the representative frames of key facial poses; and cleaning incoming frames of facial marker data using the facial pose matrix.