Related Experiment Video
Updated: Nov 24, 2025
06:32
Author Spotlight: Automated Deep Brain Stimulation for Parkinson's Disease - Exploring the Possibilities and Challenges of Home Monitoring
Published on: July 14, 2023
1.6K
Evaluation of 3D Markerless Motion Capture Accuracy Using OpenPose With Multiple Video Cameras.
Nobuyasu Nakano1,2, Tetsuro Sakura3, Kazuhiro Ueda3
1Department of Life Sciences, Graduate School of Arts and Sciences, The University of Tokyo, Tokyo, Japan.
Frontiers in Sports and Active Living
|December 21, 2020
Summary
Markerless motion capture using OpenPose offers a viable, accurate alternative for human movement science. With potential for <30 mm error, this technique simplifies motor performance evaluation.
Area of Science:
- Human Movement Science
- Biomechanics
- Sports Technology
Background:
- Marker-based motion capture is standard but complex.
- A need exists for simpler, accurate markerless systems.
- Evaluating motor performance requires precise movement analysis.
Purpose of the Study:
- Develop and validate a 3D markerless motion capture system using OpenPose.
- Compare the accuracy of OpenPose-based system against optical marker-based systems.
- Assess the system's suitability for human movement science applications.
Main Methods:
- Utilized OpenPose software with multiple synchronized cameras for 3D pose estimation.
- Recorded participants performing walking, jumping, and throwing tasks.
- Compared joint position data from markerless and marker-based systems calculating Mean Absolute Error (MAE).
Main Results:
- Markerless motion capture qualitatively reproduced human movements accurately.
- 80% of joint position errors were below 30 mm.
- 10% of errors exceeded 40 mm due to tracking failures.
Conclusions:
- OpenPose-based markerless motion capture shows promise for human movement analysis.
- Accuracy can reach below 30 mm with improved tracking algorithms.
- This technology can enhance motor performance evaluation in scientific research.
