Computational Forensic Facial Reconstruction

Rapid  developments  in  three-dimensional  (3D)  digitised  image  capture,  computer  visualisation modelling  and  animation  have  begun  make  inroads  into  many  areas  of  the  forensic  sciences, including the rather conservative specialisation of forensic pathology. This has been the result of an interdisciplinary  collaboration  between  forensic  pathologists,  biological  anthropologists  and computer  scientists.  For  example  (Figure  1),  we  have  modelled  the  track  of  the  blade  in  a  sharp force  trauma  or  stabbing  incident  and  been  able  to  exclude  certain  body  postures  as  having occurred at the time of the injury (March et al.2003). Forensic  facial  reconstruction  serves  to  illustrate  one  aspect  of  the  results  of  these  collaborations, but  surveying  and  reconstruction  and  modelling  and  animation  of  accident  (Figure  2)  or  crime scenes  (Figure 3) are other fields in which 3D computerised methods are  gradually being  adopted (Noond et  al.  2002).  Again,  visualisation  can  be  used  as  an  investigative  tool—in  comparing scenarios based on conflicting witness statements for example. I would like to briefly review 3D forensic facial reconstruction, describing the traditional method, and outlining the novel approaches being adopted in computerised 3D forensic facial reconstruction by several research groups.