GPlacencia.com

Twelve law enforcement agencies in the United Kingdom are using a forensic facial composite software tool that does not require witnesses to provide detailed descriptions of suspects they can barely remember. Instead, the EFIT-V software has the witness answer questions about the age, sex, face shape, and hairstyle of the suspect to initialize the system and then produce a set of computer-generated faces. The operator of the system shows the witness the lineup and has the individual select the face that represents the best likeness of the suspect, and then variants of the face are produced and the witness chooses again from the new set of images. "Unlike traditional feature-based methods, the approach described here utilizes global, whole face, facial characteristics and allows a witness to produce plausible, photo-realistic face images in an intuitive way," according to Stuart Gibson of the University of Kent, Canterbury and colleagues. The EFIT-V system evolves the face into the likeness of the suspect. In trials, the system doubled the useful intelligence of conventional methods. The research appeared in the International Journal of Electronic Security and Digital Forensics.

(from ACM TechNews)