Estimating human motion is an area of emerging interest in instrumentation and measurement. Importantly, in such applications the physicality of the solution is paramount. Therefore, our paper proposes the SHAPE algorithm to estimate angular velocities of a moving segment from a cluster of passive markers used in a standard motion capture system known as stereophotogrammetry. We demonstrate the efficacy of our formulation with reference to the well known shape-matching algorithm in computer graphics. It is shown that our method is superior over the shape-matching algorithm.