Robust Human Pose Estimation and Action Recognition Utilizing Feature Extraction

Key Points

• This research aims to enhance human pose estimation and action recognition through the integration of inertial measurement units and monocular vision methods.
• Utilized inertial measurement units (IMUs) alongside monocular vision techniques for data collection.
• Applied feature extraction techniques to enhance the accuracy of human pose estimation.
• Conducted comparative analysis of the proposed method against existing techniques.
• Demonstrated improved accuracy in pose estimation compared to traditional methods.
• Achieved reliable action recognition with minimal latency, enhancing real-time applications.
• Highlighted the potential for integration in diverse fields, including healthcare and human-computer interaction.