Related Experiment Video
Updated: Sep 6, 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
9.1K
Combining AI Techniques for Recognizing Aerobics Sport Videos
1Zhengzhou Preschool Education College, Zhengzhou 450000, China.
Applied Bionics and Biomechanics
|June 27, 2022
Summary
This study introduces a new framework for recognizing aerobics athletes' actions using multi-feature fusion. The method enhances accuracy in identifying athlete movements and trajectories, crucial for sports analysis.
Area of Science:
- Sports Science
- Biomechanical Analysis
- Computer Vision
Background:
- Accurate recognition of aerobics athletes' actions is crucial for performance analysis and training.
- Existing methods may struggle with complex movements and subtle feature variations.
Purpose of the Study:
- To propose a multifeature fusion-supported framework for improving aerobics athletes' action recognition accuracy.
- To enhance the identification of athlete trajectories and movement similarities.
Main Methods:
- 3D peripheral structure reconstruction of aerobics footprints.
- Fuzzy feature decomposition and multi-pane fusion of footprint images.
- Wear polymorphic liquefaction for athlete trajectory similarity identification.
Main Results:
- Demonstrated superior similarity recognition of athlete footprints compared to existing methods.
- Achieved higher accuracy in behavior location and precise footprint recognition.
- Exhibited a small mean square error for aerobics movements and high recognition fidelity.
Conclusions:
- The proposed framework effectively improves aerobics action recognition accuracy and trajectory identification.
- The system offers a reliable foundation for sports science applications, including bodybuilding and potentially game analysis.
- The method addresses challenges in recognizing complex aerobics coordination with high fidelity.
