Body pose estimation in depth images for infant motion analysis

Motion analysis of infants is used for early detection of movement disorders like cerebral palsy. For the development of automated methods, capturing the infant's pose accurately is crucial. Our system for predicting 3D joint positions is based on a recently introduced pixelwise body part classifier using random ferns, to which we propose multiple enhancements. We apply a feature selection step before training random ferns to avoid the inclusion of redundant features. We introduce a kinematic chain reweighting scheme to identify and to correct misclassified pixels, and we achieve rotation invariance by performing PCA on the input depth image. The proposed methods improve pose estimation accuracy by a large margin on multiple recordings of infants. We demonstrate the suitability of the approach for motion analysis by comparing predicted knee angles to ground truth angles.

I. Introduction

Movement disorders like cerebral palsy (CP) can be detected at an early age. The current medical gold standard method for early detection of CP is the General Movements Assessment (GMA) [1], which requires a trained expert, often a doctor, to manually examine video recordings of infants to evaluate their movements. Multiple drawbacks exist for this method: it is time-consuming, it requires an expert who is repeatedly trained on the GMA, and the outcome is based on a subjective opinion.