Related Experiment Video
Updated: Aug 5, 2026
12:39
A Methodology for Capturing Joint Visual Attention Using Mobile Eye-Trackers
Published on: January 18, 2020
Attention-Based Multimodal Framework for Athlete-Performance Analysis and Rehabilitation Monitoring Using Vision and
Mohammed Alonazi1, Iqra Aijaz Abro2, Maha Abdelhaq3
1Department of Information Systems, College of Computer Engineering and Sciences, Prince Sattam bin Abdulaziz University, Al-Kharj 16273, Saudi Arabia.
Bioengineering (Basel, Switzerland)
|July 28, 2026
Summary
This study developed an AI framework using wearable sensors and computer vision for accurate human activity recognition (HAR). The multimodal approach enhances athlete performance analysis and rehabilitation monitoring by analyzing movement patterns.
Area of Science:
- Sports Science
- Rehabilitation Technology
- Artificial Intelligence
Background:
- Wearable sensors, computer vision, and AI are crucial for movement analysis in sports science and rehabilitation.
- These technologies enable biomechanical assessment, injury prevention, and training optimization.
- Intelligent multimodal sensing systems are vital for continuous evaluation of movement quality and recovery.
Purpose of the Study:
- To develop and evaluate an attention-based multimodal framework.
- Integrate wearable inertial sensing and RGB video analysis.
- Achieve robust athlete-performance assessment and rehabilitation monitoring via accurate human movement pattern recognition.
Main Methods:
- Combined inertial sensor data and RGB visual information for athlete performance analysis and rehabilitation monitoring.
- Utilized adaptive windowing for inertial signal segmentation and silhouette refinement for visual motion analysis.
- Employed multimodal feature fusion with Ranger optimization and an attention-based deep learning classifier.
Main Results:
- Achieved high accuracy scores of 88.40% (VIDIMU) and 87.96% (UTD-MHAD).
- Multimodal recognition proved more robust than single-modality solutions.
- Integration improved analysis of complex movements under varied conditions.
Conclusions:
- The multimodal framework supports intelligent athlete-performance and rehabilitation monitoring.
- Potential applications include biomechanical assessment, training monitoring, and injury-risk management.
- Highlights the integration of wearable sensing, computer vision, and AI for robust movement analysis.
