Skip navigation

Comparing individual and morphed facial composites constructed using feature-based and holistic systems

Comparing individual and morphed facial composites constructed using feature-based and holistic systems

Davis, Josh P., Solomon, Chris and Gibson, Stuart (2011) Comparing individual and morphed facial composites constructed using feature-based and holistic systems. In: 2011 AP-LS Annual Meeting and the 4th International Congress of Psychology and Law, 2-5 March 2011, Hyatt Regency Miami Hotel, Florida, USA. (Unpublished)

Full text not available from this repository.

Abstract

Facial composites produced by eyewitnesses with the assistance of the police are often a poor likeness to the offender. However, recently developed computerized systems that are designed to closely utilize the holistic processes implicated in face perception may provide a solution. Experiments are reported in which composites produced using E-FIT, a feature-based system, were compared to EFIT-V, a contemporary holistic system. No system differences in similarity to the target were found. However, morphing up to 16 composites from both systems was found to be advantageous, resulting in increased likeness ratings to the target and also increased correct naming rates.

Item Type: Conference or Conference Paper (Speech)
Additional Information: [1] Session delivered on 5th March (09:00 to 10:20) (session number 078 on the programme, appearing on page 72. *** [2] The 2011 AP-LS Annual Meeting was held in conjunction with the 4th International Congress of Psychology and Law.
Uncontrolled Keywords: facial composites, eyewitnesses, holistic processes, face perception, E-FIT(a feature-based system), EFIT-V (a contemporary holistic system)
Pre-2014 Departments: School of Health & Social Care
School of Health & Social Care > Applied Psychology Research Group
Related URLs:
Last Modified: 14 Oct 2016 09:13
Selected for GREAT 2016: None
Selected for GREAT 2017: None
Selected for GREAT 2018: None
URI: http://gala.gre.ac.uk/id/eprint/5200

Actions (login required)

View Item View Item