A Growth-Based Framework for Scaling Infant Anthropometry in Motion Capture Studies

Biomechanical analysis of infant motor behavior is complicated by rapid growth and large inter-subject variability in body size and segment proportions. We present a growth-adjusted anthropometric scaling framework that integrates Centers for Disease Control and Prevention (CDC) growth charts with three-dimensional motion capture to estimate subject-specific infant biomechanics during prone play. Nineteen infants aged 2-6 months were recorded during spontaneous prone movement. Weight-for-age percentiles were computed using CDC LMS parameters and used to infer body length from length-for-age references, ensuring internal consistency between weight and length estimates. Body segments were scaled using published infant proportions and validated against head and shoulder widths measured from motion capture. Infant head and trunk proportions were substantially larger than adult values, whereas limb proportions were similar. Estimated and measured dimensions of head and trunk width showed good agreement, with mean root mean squared errors of approximately 3 cm. Adult-based scaling produced head and trunk length errors of up to 6 cm. This growth-adjusted framework improves anatomical fidelity and enables normalization of biomechanical measures across infants with differing growth trajectories.