Effect of Camera Configurations on Trunk and Lower-Extremity Kinematic Accuracy Using OpenCap with Lab-Based Motion Capture: A Pilot Investigation

This study aimed to explore the feasibility of using OpenCap with a lab-based multi-camera motion capture (MOCAP) system and to evaluate the influence of camera configurations on the trunk and lower-limb kinematic accuracy for athletic movements. Six top-level female football players performed drop jump (DJ) and sidecut (SC) tasks while their motion was recorded synchronously by marker- and video-based cameras. Lower-limb and trunk joint angles, as well as pelvis translations and rotations, obtained with three camera configurations—two, four, and eight cameras—were compared with the marker-based reference. Statistical parametric mapping repeated measures ANOVAs revealed significant differences in kinematic waveforms, decreasing from 16 and 14 degrees of freedom in the two-camera configuration during DJ and SC, respectively, to 9 degrees of freedom in the eight-camera configuration. Improvements in root mean square error were also observed in pelvis anterior–posterior translation and pelvis rotation in both tasks; vertical translations, right ankle dorsiflexion, and inversion in DJ; lumbar bending, and right hip rotation in SC. The result suggested that using a lab-based MOCAP system with more cameras could modestly enhance accuracy and provide several advantages, e.g., broader camera coverage. However, further investigation is needed to ensure the differences are biomechanically meaningful.