Human face anthropometric measurements using consumer depth camera

Abstract

Human face anthropometric measurements are used in forensics, orthodontics, face modelling and many other domains, wherein distance between set of facial landmarks play an important role to make inferences. 3D facial data captured using specialized acquisition methods can be used to reduce the time and tedious task involved in order to compute these measurements. The proposed method is developed to compute fifteen canonical linear measurements between facial landmarks using Kinect camera. Results obtained from this system are compared with the traditional method of measurement using digital Vernier caliper. The experimental results indicate that measurements using RGB-D data obtained from Kinect are good enough for a quick preliminary assessment of the subject as compared to traditional method.