Head measurements from 3D point clouds

Head measurements are widely used in different fields as ergonomics, medicine and acoustics. In acoustics they are useful to create 3D virtual auditoriums, since the manipulation of the Head Related Transfer Function (HRTF) allows to virtually place sound sources. Despite the HRTF dependents on the physical characteristics of each person, a generic HRTF is frequently adopted in commercial systems because the user's anthropometric data is not always available. In this paper we present a scheme to calculate several head dimensions required to individualize the transfer function, however they can also be utilized in other applications. The measurements are calculated automatically on 3D models of the head acquired with an inexpensive RGBD sensor. Experiments performed with 80 point clouds of 20 subjects corroborate the potential of the proposed algorithm.