The Gaitprint: Identifying Individuals by Their Running Style
Christian Weich1, Manfred M Vieten1
1Sports Science, University of Konstanz, 78464 Konstanz, Germany.
Sensors (Basel, Switzerland)
|July 12, 2020
Summary
Researchers developed a new method using micro-electro-mechanical-system (MEMS) sensors to identify athletes by their unique running style. This gaitprint analysis achieved 99% accuracy, offering new insights into motion analysis and personal identification.
Area of Science:
- Biomechanics
- Sensor Technology
- Data Analysis
Background:
- Quantifying human movement is crucial in sports science.
- Portable micro-electro-mechanical-system (MEMS) sensors enable precise motion capture.
- Identifying individuals by movement patterns is an emerging field.
Purpose of the Study:
- To develop and validate a method for identifying athletes based on their running style.
- To create an individual 'gaitprint' using sensor data.
- To assess the sensitivity and accuracy of the proposed recognition algorithm.
Main Methods:
- Collected movement data from 30 athletes over three 20-minute running sessions.
- Utilized a novel analysis method based on limit-cycle attractors.
- Applied a recognition algorithm to differentiate individual running styles.
Main Results:
- Achieved a 99% detection rate for identifying individual athletes.
- Reported a false identification probability of only 0.28%.
- Demonstrated that running style variations are individual modifications of a general running pattern.
Conclusions:
- The developed method offers a highly sensitive approach for athlete recognition based on running style.
- Findings suggest potential applications in sports performance assessment and broader personal identification.
- This research opens new avenues for analyzing general human motion and its application in areas like e-sports.
