Burger menu

Impact of Aging on Face Recognition: Studies, Databases and Recognition Techniques

Facial aging, while predetermined by various factors, can drastically impact a facial recognition system

General Overview

Aging is a natural physiological process that occurs in a human organism during its growth. Face is a region where aging effects are more apparent when compared to rest of the body. Apart from natural growth, the facial structure can also be impacted by age-related weight gain or loss, dietary habits, location and climate, smoking or drinking alcohol, substance abuse, sunlight exposure, hormonal imbalance, loss of teeth, and other reasons. In facial anti-spoofing, and its types, countermeasures, and challenges, aging is seen as a crucial factor capable of decreasing recognition accuracy.

A series of photos demonstrating face aging of a single person
A series of photos demonstrating face aging of a single person

This creates an obstacle for facial recognition systems, as the human face has the lowest score in terms of permanence — an essential biometric identifier. At the same time, facial recognition is seen as the most preferable modality because of its contactless, reliable and un-intrusive nature if working in the passive liveness regime.

Ranking of the biometric recognition methods or modalities in 2020
Ranking of the biometric modalities in 2020

In light of the challenges posed by aging — including a possible demographic bias in facial recognition — a need for robust algorithms, that can identify a person despite visual changes in their appearance, has emerged. A number of approaches have been proposed along with datasets and experiments to address this issue.

Statistics & General Information

An extensive research conducted at the Michigan State University involved a face recognition algorithm and two mugshot databases to identify any relationship between facial aging and recognition accuracy. The results showed that a time interval of 6 years is optimal for a system to identify an individual: the recognition accuracy rate was 99%.

Front and mugshots of men from the NIST Database used for facial recognition training with aging affects
Sample from the NIST Special Database 18

However, the success rate of recognition can vary from person to person, as genetics, lifestyle habits, and other factors can stimulate or impede facial aging. Accuracy rates would also diminish if test images offered for recognition were captured over a 6-year interval. Interestingly, criminal mugshot databases are a favorable source of sample data: they have a large selection of photos, subjects and time intervals.

Another research at Purdue University demonstrated that False Rejection Rate (FRR) had a persistent growth if a face recognition algorithm was presented with photos that had a time gap among them.

Age-invariant Face Recognition Methods

Age-invariant face recognition refers to techniques that are capable of maintaining accuracy despite noticeable appearance alterations caused by aging. They are categorized as a) Generative — they simulate face changes caused by aging b) Discriminative — they focus on specific features that stay unchanged over time c) Deep learning-based — they involve neural networks.

Generative Methods

A face anti spoofing and recognition method proposed comprises generative and discriminative aging models. Discriminative model focuses on notable facial features, which make recognition possible despite time lapses. And the generative model is responsible for learning the parametric aging model in the 3D domain to synthesize images where the potential age gap is being removed. (3D modeling fits for detecting aging patterns as aging is a three-dimensional phenomenon.)

images of people at different ages used in Aging modelling for facial recognition
Aging modelling (described above) increases matching accuracy for age-gap image recognition

An alternative method suggests recreating 3D faces from 2D photos to obtain embracive craniofacial data. It includes stages such as 2D facial feature point detection with the Active Appearance Model (AAM), 3D model fitting through Blanz and Vetter’s model and 3D facial shape represented through eigenvectors, Tikhonov regularization controlling the Mahalanobis distance, pose correction, shape and aging texture patterns, and finally aging simulation. The solution was tested with three datasets — FG-NET, MORPH and Browns — showing improved performance when recognizing age-affected images.

A man's facial features are modelled using Active Appearance model to detect facial data for recognition
Example of AAM
2D facial features of a person are fitted on a 3D model for reconstructing the real face for recognition
3D model fitting for reconstructing the 3D shape
The different steps in creating of a 3D aging model for facial recognition
Creation of a 3D aging model

An alternative method of face liveness detection is based on the Temporal Non-Volume Preserving (TNVP) transformation, which is capable of breaking down the entire aging process into temporally shorter phases. It’s capable of registering non-linear age variations and modeling a smooth synthesis in age progression and can be used in both passive and active liveness detection systems.

Non-Generative Methods

To an extent, facial mark analysis can be attributed to the non-generative approaches. It focuses on the local facial marks that retain unique characteristics and have a certain degree of permanence: freckles, lip creases, birthmarks, scars, and so on. They are frequently used as identifiers in forensics. However, plastic surgery or makeup can undermine recognition.

A woman's facial marks like freckles, moles and wrinkles can aid in facial recognition
Facial marks

Another method suggests usage of the Gradient Orientation Pyramid (GOP), which can recreate features that are immune to age-induced alteration. It also employs Support Vector Machines (SVM) used for calculating image difference pairs from the cosine between gradient orientations at all scales.

An alternative approach dubbed Hidden factor analysis implies that the face can be divided into two regions: the first region represents the facial ‘fundament’, which remains unchanged for the most part during a person's life. The second region represents the combined aging effects that occur along with the anatomical growth. Expectation–Maximization procedure is used for estimating the latent factors.

A demography-based study has also been suggested. It assumes that facial aging can be better understood through the prism of demographic estimation of the human face. The study focuses on facial symmetry, which plays a pivotal role in the aging mechanism.

Deep Neural Network Approaches

Convolutional Neural Networks are proposed for the age-invariant face recognition. One of the concepts directly advises training a CNN to detect jointly features, distances and thresholds that occur together with age-related alterations. A different method is a set-based approach, in which photos of a certain subject are put in a set of images. Then, this set is compared to image sets of other individuals, while feature extraction is conducted with a CNN.

Finally, it is recommended to fuse generative and discriminative approaches. That implies that distinctive features of age-invariant and age-sensitive regions should be separated with a CNN. Then, pixel mean vector, local binary patterns, generative model, and bridged denoising auto-encoders are used to process the mentioned features.


Datasets used for age-invariant face recognition are longitudinal. That means they include an assortment of samples collected over a long time interval from the same group of subjects. Forensic mugshot datasets are often the primary data source in this case.


MORPH is, by far, the largest longitudinal dataset with 400,000+ sample images obtained from 70,000 people. It’s periodically renewed and also supported with the metadata for each individual: gender, weight, height, etc. There are two versions: corporate and academic with 50,000+ samples from 13,000+ subjects and with the biggest time interval between photos being 1,681 days (4+ years).

Mugshots of African American woman taken over a few years can aid in aging training of facial recognition
MORPH dataset sample


FG-NET offers 1,002 images taken from 82 people. Apart from austere sample variety, it also offers images of average quality at best, which can be a favorable option for training a solution that is fit for ‘uncontrolled environments’. Currently, it may be unavailable from the official distributor. Alternative download links are here and here (courtesy of Yanwei Fu).

A woman's image at different ages from FG-NET samples for facial recognition training
FG-NET samples


LEO_LS includes 31,852 images of 5;636 people with the biggest time interval of 8 years. PCSO_S contains 147,784 operational mugshots captured from 18,007 repeat criminal offenders. The photos were shot in the time period from1994 to 2010. Currently, both datasets seem to be unavailable.

Images of people at different ages from PCSO_S database for facial recognition training
PCSO_S database sample

NIST Special Database 18

This dataset is not exactly longitudinal. However, it features 3,248 images taken from 1,573 individuals posing for mugshots in front view and profile. Potentially, the dataset can be used for training a generative face recognition method.

Solutions & Experiments

A number of pre-trained CNNs — Alex-Net, GoogleNet, Inception V3, ResNet50 and SqueezeNet — were used for testing an algorithm capable of feature extraction for age-invariant face recognition. As it turned out, a descriptor Fc7 supported with an SVM classifier achieved a 98.21% accuracy rate. AlexNet, also combined with an SVM classifier, showed the highest performance score with a 98.21% accuracy rate.

Performance results of CNN-based feature extraction testing for facial recognition
CNN-based feature extraction testing results


Does aging affect face recognition?

Aging can impede facial recognition under certain conditions.

Impact of facial aging on face recognition is viable as human face is subject to natural changes, as well as those caused by lifestyle habits, health condition, climate, and other factors. An experiment was conducted at Michigan State University, which showed that human face images stay recognizable to a biometric system if the time interval between them isn’t more than 6 years.

To address the issue, a number of facial aging datasets and solutions are proposed. Among the recognition methods are usage of generative and discriminative aging models, extraction of embracive craniofacial data from recreated 3D images, and others.

Which biometric identifier is most prone to aging?

Face is feasibly susceptible to aging.

Face is considered the most aging-sensitive biometric trait. While it’s affected by the natural process of growth, it’s also influenced by the climate conditions, health habits, gastronomic preferences, lifestyle, stress, and so on.

Other traits are less affected by aging. For example, human fingerprints stay permanent for the entirety of life with a few minor changes that cannot thwart biometric recognition and antispoofing. Iris is believed to be stable, even though it may show genuine match score distribution declining over the course of time. Aging also affects the larynx and vocal cords altering the voice.

What are the main age-invariant face recognition methods?

There’s a group of methods to recognize aging faces.

A set of techniques for aging face recognition is proposed — age-invariant methods. They are separated into Generative and Non-generative groups. The first group implies analysis of discriminative and aging models. The former pays attention to the facial features that stay permanent over the time, like facial contour. The latter learns the parametric aging model, synthesizes images and then removes aging from them.

Non-generative methods include facial mark analysis, which explores unique traits that stay more or less unchanged: freckles, birthmarks, and others. This method may fail or be spoofed with a surgery.

Are there any databases of the same people of different ages?

A limited number of facial aging datasets exists, as of today.

Facial aging databases are presented in 4 main examples: MORPH, FG-NET, PCSO_S, and LEO_LS.

MORPH is an extensive longitudinal dataset that contains more than 400,000 photos retrieved from 70,000 individuals. Its main advantages are periodic updates that introduce new samples and detailed metadata accompanying each image.

FG-NET is a humbler specimen with 1,002 photos provided by 82 volunteers. The photo quality is non-professional, so the dataset can be used for rehearsing ‘unconstrained scenarios’ that biometric solutions often deal with. PCSO_S contains a large collection of mugshots and LEO_LS offers 31,852 images of 5;636 people.

What are the main datasets for age-invariant face recognition?

Four datasets are used for training age-invariant face recognition.

Four primary facial aging datasets explore age-induced effects on the human face. The biggest example is the MORPH database that provides 400,000+ images featuring 70,000 volunteers. While being licensable, it periodically updates its sample collection and specifies each person’s age, height, gender, weight, and other metadata that can be used in biometric identification.

FG-NET is a free database with 1,002 images of average quality featuring 82 volunteers. LEO_LS comprises 31,852 samples of 5;636 people (the largest time-lapse is 8 years.) PCSO_S has 147,784 mugshots of repeat criminal offenders.


  1. Factors influencing face aging. Literature review
  2. A series of photos demonstrating face aging of a single person
  3. Year in Review: COVID Shakes Up the Top Biometric Modality Rankings
  4. Aging faces could increase security risks
  5. Sample from the NIST Special Database 18
  6. Aging effects in automated face recognition
  7. Face recognition: Some challenges in forensics
  8. Age-Invariant Face Recognition
  9. Active Appearance Model by Wikipedia
  10. A Morphable Model for the Synthesis of 3D Faces
  11. Tikhonov Regularization and ERM
  12. Mahalanobis Distance – Understanding the math with examples (python)
  13. FG-NET
  14. MORPH
  15. Active Appearance Models in C++ (Paamela)
  16. Deep-learning based descriptors in application to aging problem in face recognition
  17. Temporal Non-Volume Preserving Approach to Facial Age-Progression and Age-Invariant Face Recognition
  18. Detection and matching of facial marks in face images
  19. Examples of facial marks Scar, mole, and freckles etc.
  20. Gabor-based gradient orientation pyramid for kinship verification under uncontrolled environments
  21. A Gentle Introduction to Expectation-Maximization (EM Algorithm)
  22. MORPH Facial Recognition Database
  23. academic
  24. MORPH dataset sample
  25. FG-NET dataset by Yanwei Fu
  26. PCSO_S database sample
  27. NIST Special Database 18
  28. AlexNet: The Architecture that Challenged CNNs
Avatar Antispoofing


Editors at Antispoofing Wiki thoroughly review all featured materials before publishing to ensure accuracy and relevance.