Introduction

Facial images are one of the most valuable pieces of evidence for identification and it is possible in a very high proportion of cases wherein subjects recorded in photographs or on videotapes needs to be identified by means of comparison with images of persons of known identity (Stavrianos et al., 2012). Utilizing images to explain and record findings for forensic and diagnostic applications is known as forensic imaging (Zhang, 1). "Facial imaging" approaches are those that contribute in human identification by analysing or providing facial graphics. Thus, facial imaging includes approximation of the face and photographic superimposition, as well as include age progression and regression, the formation of facial graphics from eyewitness memory (such as composites and sketches), facial depiction, face mapping, and newly invented techniques of so-called "molecular photo fitting (Stephan et al., 2018)."

Methods for facial imaging are essential in many situations. In other circumstances, they offer standard police investigation techniques and crucial evidence in cases that would not have been solved otherwise. The majority of the aforementioned techniques—with the exception of automatic facial recognition systems—succeed by getting the public's and the media's attention on the rebuilt image of a face. This is true even if there are times when the successful outcome is not directly related to the picture of the face (e.g. the Instead of the estimated facial morphology itself, recognition of other things such as personal clothing like ties, spectacles, hats, necklaces, or shirts exhibited with the face (Stephan et al., 2018).

It is advantageous to maximize their capacity to identify people based on facial morphology. Applying science to test or enhance approaches used in the field and in the lab are beneficial. As Regardless of how lofty they may seem, advancements that produce any technique to be so trustworthy that it can be applied for Identification methods used alone are optimum, and shouldn't be completely disregarded in some situations. This provides maybe novel techniques and Possibilities for identification and distinctive opportunities for forensic anthropologists in training increase and contribute to forensic anthropology input. As more forensic professionals become aware of its benefits, the application space for forensic imaging has also expanded. This method has also been applied in other forensic fields, such as forensic anthropology, forensic odontology, forensic ballistics, and animal forensics, among others, in addition to forensic pathology (Zhang, 1). FaceGAN is a network type that consists of a generator, a discriminator, a deep face recognition model, and a deep age estimation model. The discriminator is taught in a multi-task learning environment, distinguishing between genuine and synthetic samples and categorizing real and synthetic photos into matching age ranges at the same time. Numerous tests using the UTKFace dataset have shown that Face- GAN can produce photo-realistic face images at various ages for both face ageing and rejuvenation while effectively maintaining individual identification (Zeng et al., 302).