Abstract—Face verification in the presence of age progression is an important problem that has not been widely addressed. In this paper, we study the problem by designing and evaluating discriminative approaches. These directly tackle verification tasks without explicit age modeling, which is a hard problem by itself. First, we find that the gradient orientation (GO), after discarding magnitude information, provides a simple but effective representation for this problem. This representation is further improved when hierarchical information is used, which results in the use of the gradient orientation pyramid (GOP). When combined with a support vector machine (SVM) GOP demonstrates excellent performance in all our experiments, in comparison with seven different approaches including two commercial systems. Our experiments are conducted on the FGnet dataset and two large passport datasets, one of them being the largest ever reported for recognition tasks. Second, taking advantage of these datasets, we empirically study how age gaps and related issues (including image quality, spectacles, and facial hair) affect recognition algorithms. We found surprisingly that the added difficulty of verification produced by age gaps becomes saturated after the gap is larger than four years, for gaps of up to ten years. In addition, we find that image quality and eyewear present more of a challenge than facial hair. Index Terms—Face verification, age progression, gradient orientation pyramid, support vector machine A. Background I.

Estimation of a person’s age from the facial image has many applications, ranging from biometrics and access control to cosmetics and entertainment. Many image-based methods have been proposed for this problem. In this paper, we propose a method for the use of dynamic features in age estimation, and show that 1) the temporal dynamics of facial features can be used to improve imagebased age estimation; 2) considered alone, static image-based features are more accurate than dynamic features. We have collected and annotated an extensive database of face videos from 400 subjects with an age range between 8 and 76, which allows us to extensively analyze the relevant aspects of the problem. The proposed system, which fuses facial appearance and expression dynamics, performs with a mean absolute error of 4.81 (±4.87) years. This represents a significant improvement of accuracy in comparison to the sole use of appearance-based features.

Analysis of face images has been the topic of in-depth research with wide spread applications. Face recognition, verification, age progression studies are some of the topics under study. In order to facilitate comparison and benchmarking of different approaches, various datasets have been released. For the specific topics of face verification with age progression, aging pattern extraction and age estimation, only two public datasets are currently available. The FGNET and MORPH datasets contain a large number of subjects, but only a few images are available for each subject. We present a new dataset, VADANA, which complements them by providing a large number of high quality digital images for each subject within and across ages (depth vs. breadth). It provides the largest number of intrapersonal pairs, essential for better training and testing. The images also offer a natural range of pose, expression and illumination variation. A parallel version with aligned faces is also created. Additionally, we provide relationships between subjects. We demonstrate the difference and difficulty of VADANA by testing with state-of-the-art algorithms. Our findings from experiments show how VADANA can aid further research on different types of verification algorithms. The variety of characteristics our data offers facilitate testing and benchmarking of other facial analysis algorithms. 1.