Action Quality Evaluation Using Motion Signal Separation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing portable electronic devices fail to provide actionable and user-friendly performance metrics for fitness activities, as the performance parameters are often abstract and do not meet user needs.
Innovation Solution
A method and apparatus for action quality evaluation that determines action evaluation metrics such as consistency, stability, continuity, variability, and force control based on motion data, independent of exercise type, using singular value decomposition and data reconstruction to isolate primary and secondary motion signals.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If performance parameters are measured using conventional methods, then measurement coverage is achieved, but the parameters are overly abstract and not user-friendly
Solution Approach 1:
The patent segments performance parameters into multiple dimensions: action quality metrics (consistency, stability, continuity, variability, force control) and exercise type metrics. This segmentation transforms abstract performance data into specific, actionable categories that users can understand and improve upon individually.
Solution Approach 2:
The patent changes the parameter representation from conventional abstract performance metrics to specific action quality indicators. By defining new parameters such as amplitude consistency, completion time consistency, jitter degree, and force exertion condition, the system transforms unusable abstract data into actionable performance feedback.
2Measurement precision
If data reconstruction processing is performed to eliminate exercise type influence, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent extracts and eliminates the exercise type component from motion data through data reconstruction processing. By separating the exercise type influence from the action quality signals, the system achieves precise action quality measurement independent of the specific exercise being performed, while keeping the core evaluation logic unified.
Solution Approach 2:
The patent creates a universal evaluation system that can assess action quality across different exercise types using the same algorithm and model. The data reconstruction processing enables a single unified framework to handle multiple exercise types, reducing the need for exercise-specific evaluation logic despite the increased processing complexity.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The embodiments of the present disclosure provide methods and apparatuses for action quality evaluation, storage media and electronic devices, where the method for action quality evaluation includes: acquiring motion data of a user collected during a fitness session; determining a value of at least one movement evaluation metric of the user based on the motion data; and determining an action quality evaluation result of the user based on the value of the at least one movement evaluation metric. With the scheme for action quality evaluation provided by the embodiments of the present disclosure, the action quality evaluation result determined based on the value of one or more action evaluation metrics can be utilized to characterize the action quality of the user, which is easier for the user to understand, and thus meets the needs of the user.