Abstract

The goal of craniofacial reconstruction is to infer from a person’s skull the shape of their face. It is frequently used in forensic medicine, archaeology, cosmetic surgery, and other fields. However, the assessment of craniofacial reconstruction receives minimal attention. The geodesic network-based reconstruction of the craniofacial faces is evaluated both globally and locally using an objective method proposed in this paper. First off, the geodesic networks of the original face and the reconstructed craniofacial face are created using geodesics and geodesics, respectively, whose intersections constitute network vertices. Then, the weighted average of the shape index values in a neighbourhood is defined as the feature of each network vertex, and the absolute value of the correlation coefficient of the features of all related geodesic network vertices between two models is considered as the holistic similarity. Additionally, the local similarity is calculated for each of the six subareas of each model’s geodesic network, which are the forehead, eyes,nose, mouth, cheeks, and chin. The subjective judgement received from 35 people in five groups is broadly similar with the evaluation made by our method, according to experiments using 100 pairs of reconstructed craniofacial faces and their corresponding original faces.