Facial Masks and Soft-Biometrics: Leveraging Face Recognition CNNs for Age and Gender Prediction on Mobile Ocular Images

TL;DR

This paper explores using lightweight CNNs, pre-trained on face recognition tasks, to estimate age and gender from selfie ocular images captured with smartphones, addressing challenges posed by face masks and mobile device limitations.

Contribution

It adapts and evaluates lightweight CNN architectures pre-trained on large datasets for soft-biometrics prediction from mobile ocular images, demonstrating improved accuracy with face recognition fine-tuning.

Findings

Pre-trained face recognition models enhance age and gender prediction accuracy.

Fine-tuning on face recognition datasets improves soft-biometrics estimation.

Lightweight CNNs are feasible for mobile-based biometric tasks.

Abstract

We address the use of selfie ocular images captured with smartphones to estimate age and gender. Partial face occlusion has become an issue due to the mandatory use of face masks. Also, the use of mobile devices has exploded, with the pandemic further accelerating the migration to digital services. However, state-of-the-art solutions in related tasks such as identity or expression recognition employ large Convolutional Neural Networks, whose use in mobile devices is infeasible due to hardware limitations and size restrictions of downloadable applications. To counteract this, we adapt two existing lightweight CNNs proposed in the context of the ImageNet Challenge, and two additional architectures proposed for mobile face recognition. Since datasets for soft-biometrics prediction using selfie images are limited, we counteract over-fitting by using networks pre-trained on ImageNet. Furthermore, some networks are further pre-trained for face recognition, for which very large training databases are available. Since both tasks employ similar input data, we hypothesize that such strategy can be beneficial for soft-biometrics estimation. A comprehensive study of the effects of different pre-training over the employed architectures is carried out, showing that, in most cases, a better accuracy is obtained after the networks have been fine-tuned for face recognition.