Vision-based motion capture system for tracking crawling motions of infants

摘要

This work is part of an ongoing effort to explore a new mode of intervention for children at risk of cerebral palsy. The intervention uses an autonomous mobile robot that recognizes crawling like movements to respond to the infant. We propose a new system to capture crawling motions of infants. Our methodology involves placing planar marker patterns on the body and tracking them using cameras. A single camera can be used to track the markers in 3D. Such markers are generally used in augmented reality (AR) applications but are relatively unheard of for motion capture. In this paper, we present the system design and some preliminary test results. The test results indicate that it is possible to detect displacement typical of crawling with markers as small as 27 mm on a side.