A 3D Study of Facial Mimicry: A New Approach for the Identification of Missing Children

After attending this presentation, attendees will better understand how to improve the possibility of facial recognition of missing subjects by exploiting the memory for facial expressions. This presentation will impact the forensic science community by highlighting an important 3D facial analytical method that can extract invariant features from the 3D face template. The results could be very helpful in the interpretation of 2D images, given that the 3D verification of the mimicry in childhood/adolescent samples revealed the persistence of highly individualizing facial expressions. The assessment of facial morphology is mainly used in the facial identification of missing subjects when photographs and videos are available; however, in everyday life, faces are in constant motion and often a person is recognized for distinct facial expressions. Facial morphology also is altered with age and environmental conditions. Even a simple instantaneous contact captures the image of a moving face. For this reason, researchers have begun to consider how facial motion affects memory for faces. The face provides a distinctive indication of the individual identity and previous studies demonstrated that memory for a moving face can help personal recognition.1 Moreover, in previous studies conducted with 3D methods, adults undergoing reconstructive facial surgery retained the same mimic movements one year after surgery, despite the changes to the underlying bones. This study proposes a method for personal identification of missing children, at a period of time after their disappearance, using 3D analysis of facial movements. New technologies enable presentation and manipulation of digital videos, techniques that are fundamental in the efforts to identify subjects, including minors. The identification of facial expressions can implement the effectiveness of recognition of missing persons. Three boys and three girls aged 7-14 years of age in T1 and 14-21 years of age in T2 were recruited for the study. Facial surface data were acquired by a 3D laser scanner (Cyberware 3030RGB™) and transferred to Geomagic Design™ X software for data elaboration. In previous studies, the development of a specific protocol resulted in a mean scanning error of 1mm–1.2mm and a recording error of 0.3mm–0.4 mm on repeated scans of human subjects. The patients were registered in a resting position and after movements of the upper, middle, and lower area of the face. They were frowning, grimacing, smiling, and pursing their lips. To study the modifications of these faces during growth, an anthropometric evaluation was conducted using 3D points to discern distance measurements in a resting position. Statistically significant results were obtained in longitudinal and oblique 3D measurements (p<0.05). In T1 children, the motion was easier and spontaneous in the lower area of the face, while the movements of the upper area were not intuitive. The amplitude of the movements of the lower area of the face was augmented with age. In T2, spontaneous movements in the upper area of the face appeared more evident (i.e., grimace). This is consistent with the literature. Two observers separately evaluated the soft tissue displacements during the facial movements of T1 and T2. The 3D morphological study of the mobility of the soft tissues of the lower part of the face of the subjects demonstrated that the same parts of the soft tissue were involved in the same areas of the face. A unique and characteristic facial morphology was generated, different in every subject, but similar and consistent in the subject during repeated trials. In particular, the mouth, cheeks, and nosewing presented the same morphological displacement, and nose-lip sulks and asymmetries of the lower area of face were consistent over time. The maxillafacial bone structures, especially in children, vary significantly over time, but facial movements do not.2 The 3D technology is useful for facial recognition. Nevertheless, facial modifications may pose problems for automatic individual identification and recognition, and more work needs to be conductede on a larger sample, including more variations in age categories. Reference(s): 1. Sforza C., de Menezes M., Ferrario V.F. Soft- and hard-tissue facial anthropometry in three dimensions: what’s new. J Anthropol Sci Vol.91 (2013) 159-184. 2. Koudelová J., Dupej J., Bruzek J., et al. Modelling of facial growth in Czech children based on longitudinal data: Age progression from 12 to 15 years using 3D surface models. Forensic Sci Int. 248 (2015) 33-40.