|
V8 N2 Paper 1
|
Annals of the MS in Computer Science and Information Systems at
UNC Wilmington
|
Fall 2014
|
An Evaluation of Longitudinal Face Recognition Performance Throughout the Growth and Development Stages
Michael Sodomsky
Committee
Abstract
Face recognition is the process of identifying or verifying the identity of an individual based
upon the facial features of the individual. Automated face recognition has received a lot of attention over the last 15 years. The vast majority of research and development in automated face recognition has
focused on adults. This work will evaluate the performance of automated face recognition of non-adults.
Non-adults are those persons younger than 18 years of age. The changes of the cranium and thus, the outward appearance of the face, alters significantly from birth to approximately age 16. This period is composed of the growth and development stages. This work will investigate standard and commercial face recognition systems against a set of longitudinal images during the growth and development stages. It is an established fact that face recognition performance degrades against aging, i.e. when the enrolled face is temporally displaced by the probe face. This fact has been established in adult faces; however, the question has not been adequately investigated for non-adult periods, i.e. children from birth to adulthood.
Our research has established this is a challenging problem.
download
(pdf)
Recommended Citation:
Sodomsky, M., Ricanek, K, Hua, L., Cummings, J. (2014) An Evaluation of Longitudinal Face Recognition Performance Throughout the Growth and Development Stages. Annals of the Master of Science in Computer Science and Information Systems at UNC Wilmington, 8(2) paper 1. http://csbapp.uncw.edu/data/mscsis/full.aspx.
|
V8 N2 Paper 1
|
Annals of the MS in Computer Science and Information Systems at
UNC Wilmington
|
Fall 2014
|
