Inter-session repeatability of markerless motion capture gait kinematics

Area of Science:

• Biomechanics
• Human Movement Analysis
• Motion Capture Technology

Background:

• Marker-based motion capture is the standard for gait analysis but has limitations.
• Markerless motion capture offers practical advantages for collecting gait kinematics.
• Reliability of markerless systems is crucial for clinical adoption.

Purpose of the Study:

• To assess the reliability of lower limb gait kinematics derived from markerless motion capture.
• To compare markerless system reliability against established protocols for marker-based systems.
• To determine the suitability of markerless motion capture for clinical gait analysis.

Main Methods:

• Utilized an established reliability testing protocol with eight healthy adults.
• Recorded over-ground walking trials using eight synchronized cameras.
• Computed 3D lower limb joint angles from video-based pose estimates.

Main Results:

• Inter-trial variability averaged 2.5° across joint angles, slightly higher than some marker-based studies.
• Inter-session variability averaged 2.8°, demonstrating lower values than previously reported.
• Variability ratios averaged 1.1, indicating minimal impact of multi-session protocols on total variability.

Conclusions:

• Markerless motion capture provides reliable measurements of lower limb gait kinematics.
• The system demonstrates practical benefits and potential for clinical application in gait analysis.
• This technology advances the field of human movement analysis and clinical biomechanics.