Model-based approach for human kinematics reconstruction from markerless and marker-based motion analysis systems.

Area of Science:

• Biomechanics
• Human Motion Analysis
• Musculoskeletal Modeling

Background:

• Existing musculoskeletal models lack integration of functional joint behavior, limiting accurate human motion replication.
• Current kinematic models often fail to represent individual human movement patterns effectively.
• Accurate validation of joint kinematics is crucial for meaningful interpretation of muscle behavior in motion analysis.

Purpose of the Study:

• To present a model-based method for fusing accurate joint kinematic information with motion analysis data.
• To enhance the understanding of relationships between muscle behavior and motion data.
• To validate the physiological acceptability of reconstructed human kinematics.

Main Methods:

• Developed a model-based approach (MBA) for human motion data reconstruction.
• Utilized a scalable registration method to combine physiological joint kinematics with limb segment poses.
• Integrated motion data from marker-based stereophotogrammetry (MBS) and markerless single-camera (MLS) systems.

Main Results:

• Demonstrated that the model-based MBS and MLS methods yield physiologically acceptable human kinematics.
• Showcased the successful fusion of accurate joint kinematic information with motion capture data.
• Validated the model-based approach for reconstructing individual human motion.

Conclusions:

• The proposed model-based method enables accurate reconstruction of human kinematics.
• This approach improves the interpretation of muscle behavior by ensuring validated joint kinematics.
• The method is suitable for further exploitation in musculoskeletal modeling and analysis.