3D Geometric Human Face Modeling

H. Ben Mansour , Richard A Fournier (1) , A. A. Nait-Ali (1) , A. Ben Slimane

1 SYNAPSE

In this paper we propose a new geometric model of 3D human face; this method relies on the analysis of a set of 3D facial data which are obtained by using 3D laser scanner. Our analysis of the different features points coordinates reveals that the 3D Euclidian distances in 3D scanned face are proportional to the horizontal distance between two eyes. The proposed approach has been obtained after being tested in more than 14 face scans. The set of faces scans which are used for experimental experience have different ages, different sex and also they are from different regions. Experimental results show that the proposed enhancement of such geometric model will later be used to improve not only the 3D face scanning process by knowing the location of the most features from eyebrows, nose and lips, then it is not necessary to scan all the face for recognition, but also involves to perform the extraction and the regeneration of significant facial features. Therefore, it has been found that this model can be used in biometric applications, virtual reality applications, facial expression simulation and computer based plastic surgery application.

Mots clés: 3D human face; Geometric model; Euclidian distances; scanning process; features points

Domaines: Automatique; Automatique / Robotique

H. Ben Mansour, Richard A Fournier, A. A. Nait-Ali, A. Ben Slimane. 3D Geometric Human Face Modeling. Proc. of the International Conference on Human-Machine Interaction, HMI 2013, Mar 2013, Chennai, India. pp.183-187. ⟨hal-01569681⟩