Centre of Mass Kinematic Analysis for Balance Assessment Using a Markerless Pose Estimation Device
摘要
In recent years, markerless optical systems have emerged as a cost-effective and non-invasive alternative to conventional marker-based systems for biomechanical analysis in sports, gait, and balance assessments. This study evaluates the performance of the Zed 2i stereoscopic camera, which uses neural networks for pose estimation. This study originated as part of a collaborative project with a medical team to objectively assess balance function in post-stroke patients using a mobile dynamometric platform. To simulate balance impairment conditions, two standing conditions were analysed: sway movements in anteroposterior and mediolateral directions. The total body centre of mass was estimated using anthropometric tables, and the centre of mass displacement and velocity were calculated in the x and y directions. Results indicate that the Zed 2i camera show promise as a tool for COM tracking during dynamic posturography.