Inproceedings
Evaluating Automated Facial Age Estimation Techniques for Digital Forensics
Contribution Summary
This paper presents an evaluation of existing automated facial age estimation techniques for digital forensics, focusing on their performance in determining ages and the influence of gender and actual age on their estimations. The study highlights the limitations of current methods, including the lack of sufficient sample images in specific age ranges. To address this issue, a dataset generator is proposed, which harnesses collections of several unbalanced datasets to form a balanced, curated dataset of digital images annotated with their corresponding age and gender. The evaluation of offline and cloud-based facial recognition models demonstrates the importance of having a viable dataset with an equally distributed number of faces at each age range and gender. The study's findings have significant implications for digital forensic investigators, who often face significant delays in the judicial process due to the lack of relevant experts and the arduous digital forensic process.
Keywords: Automated facial age estimation; Digital forensics; Facial recognition; Machine learning; Soft biometric traits; Age estimation; Digital forensic analysis; Child abuse investigations
Abstract
In today's world, closed circuit television, cellphone photographs and videos, open-source intelligence (i.e., social media and web data mining), and other sources of photographic evidence are commonly used by police forces to identify suspects and victims of both online and offline crimes. Human characteristics such as age, height, weight, gender, hair color, etc., are often used by police officers and witnesses in their description of unidentified suspects. In certain circumstances, the age of the victim can result in the determination of the crime's categorization, e.g., child abuse investigations. Various automated machine learning-based techniques have been implemented for the analysis of digital images to detect soft-biometric traits, such as age and gender, and thus aid detectives and investigators in progressing their cases. This paper documents an evaluation of existing cognitive age prediction services. The evaluative and comparative analysis of the various services was executed to identify trends and issues inherent to their performance. One significant contributing factor impeding the accurate development of the services investigated is the notable lack of sufficient sample images in specific age ranges, i.e., underage and elderly. To overcome this issue, a dataset generator was developed, which harnesses collections of several unbalanced datasets and forms a balanced, curated dataset of digital images annotated with their corresponding age and gender.
BibTeX
@inproceedings{anda2018facialageestimation,
author={Anda, Felix and Lillis, David and Le-Khac, Nhien-An and Scanlon, Mark},
title="{Evaluating Automated Facial Age Estimation Techniques for Digital Forensics}",
booktitle="{12th International Workshop on Systematic Approaches to Digital Forensics Engineering (SADFE), IEEE Security \& Privacy Workshops}",
year=2018,
month=05,
address="San Francisco, CA, USA",
doi={10.1109/SPW.2018.00028},
publisher={IEEE},
abstract="In today's world, closed circuit television, cellphone photographs and videos, open-source intelligence (i.e., social media and web data mining), and other sources of photographic evidence are commonly used by police forces to identify suspects and victims of both online and offline crimes. Human characteristics such as age, height, weight, gender, hair color, etc., are often used by police officers and witnesses in their description of unidentified suspects. In certain circumstances, the age of the victim can result in the determination of the crime's categorization, e.g., child abuse investigations. Various automated machine learning-based techniques have been implemented for the analysis of digital images to detect soft-biometric traits, such as age and gender, and thus aid detectives and investigators in progressing their cases. This paper documents an evaluation of existing cognitive age prediction services. The evaluative and comparative analysis of the various services was executed to identify trends and issues inherent to their performance. One significant contributing factor impeding the accurate development of the services investigated is the notable lack of sufficient sample images in specific age ranges, i.e., underage and elderly. To overcome this issue, a dataset generator was developed, which harnesses collections of several unbalanced datasets and forms a balanced, curated dataset of digital images annotated with their corresponding age and gender."
}