Related Experiment Video
Updated: Oct 24, 2025
08:25
Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment
Published on: May 7, 2019
9.2K
Visual Information Features and Machine Learning for Wushu Arts Tracking
1Sports Center, Xi'an Jiaotong University, Shaanxi 710049, Xi'an, China.
Journal of Healthcare Engineering
|August 16, 2021
Summary
This study introduces an improved image representation method for accurate martial arts tracking in videos. The machine learning approach utilizes the second-generation strip wave transform to overcome challenges like posture and lighting variations.
Area of Science:
- Computer Vision
- Artificial Intelligence
- Machine Learning
Background:
- Martial arts tracking is crucial for applications like video monitoring and 3D simulation.
- Challenges include significant appearance changes due to posture, clothing, and lighting variations.
- Accurate posture tracking in martial arts is a complex problem.
Purpose of the Study:
- To enhance the accuracy of martial arts tracking.
- To address the complexities arising from appearance variations in martial arts movements.
Main Methods:
- A novel image representation method for martial arts tracking is proposed.
- The method is based on the second-generation strip wave transform.
- It is applied to video martial arts tracking using machine learning techniques.
Main Results:
- The proposed method significantly improves the accuracy of martial arts posture tracking.
- It effectively handles variations in body posture, clothing, and lighting conditions.
Conclusions:
- The developed image representation method offers a robust solution for accurate martial arts tracking.
- This research contributes to advancements in computer vision and AI applications for human motion analysis.
