Vision-based movement recognition reveals badminton player footwork using deep learning and binocular positioning.

Area of Science:

• Biomechanics
• Sports Science
• Motor Control

Background:

• Effective footwork is crucial for dynamic interceptive actions in sports like badminton.
• Current methods for analyzing player movement lack automation and economic efficiency.
• Objective data on individual footwork is needed for specialized training programs.

Purpose of the Study:

• To develop an automated, efficient, and economical method for recording individual badminton player footwork.
• To utilize deep learning and binocular positioning for 3D footwork reconstruction.
• To analyze inter-individual adaptations in footwork during competitive play.

Main Methods:

• Deep learning algorithms were employed to extract 2D shoe coordinates from images.
• Binocular positioning was used to reconstruct the 3D coordinates of the players' shoes.
• The developed system achieved a final positioning accuracy of 74.7%.

Main Results:

• The study successfully reconstructed 3D footwork data for badminton players.
• Inter-individual differences in footwork strategies, including court distance and jump height, were identified.
• Quantitative data revealed personalized movement characteristics during projectile interception.

Conclusions:

• The proposed method provides a quantitative and objective approach to analyzing badminton footwork.
• This technology can significantly benefit athletes and motor control specialists by enabling data-driven, individualized training.
• The findings offer insights into optimizing footwork for enhanced performance in racket sports.