SFace: Privacy-friendly and Accurate Face Recognition using Synthetic Data

Abstract

Recent deep face recognition models proposed in the literature utilized large-scale public datasets such as MS-Celeb-1M and VGGFace2 for training very deep neural networks, achieving state-of-the-art performance on mainstream benchmarks. Recently, many of these datasets, e.g., MS-Celeb-1M and VGGFace2, are retracted due to credible privacy and ethical concerns. This motivates this work to propose and investigate the feasibility of using a privacy-friendly synthetically generated face dataset to train face recognition models. Towards this end, we utilize a class-conditional generative adversarial network to generate class labeled synthetic face images, namely SFace. To address the privacy aspect of using such data to train a face recognition model, we provide extensive evaluation experiments on the identity relation between the synthetic dataset and the original authentic dataset used to train the generative model. Our reported evaluation proved that associating an identity of the authentic dataset to one with the same class label in the synthetic dataset is hardly possible. We also propose to train face recognition on our privacyfriendly dataset, SFace, using three different learning strategies, multi-class classification, label-free knowledge transfer, and combined learning of multi-class classification and knowledge transfer. The reported evaluation results on five authentic face benchmarks demonstrated that the privacy-friendly synthetic dataset has high potential to be used for training face recognition models, achieving, for example, a verification accuracy of 91.87% on LFW using multi-class classification and 99.13% using the combined learning strategy. The training code and the synthetic face image dataset are publicly released 1 .

Quantifying the impact of image degradation on Deep Learning models in face recognition systems

“…Using deep learning. The following facial recognition models are present in this framework: Facebook DeepFace [Taigman et al 2014], DeepID [Sun et al 2014], FaceNet [Schroff et al 2015], VGG-Face [Parkhi et al 2015], Open-Face [Amos et al 2016], ArcFace [Deng et al 2019], and SFace [Boutros et al 2022]. Also included in this framework are SOTA models for face detection in images as Dlib [King 2009], MTCNN [Zhang et al 2016], and RetinaFace [Deng et al 2020].…”

Beyond Masks: On the Generalization of Masked Face Recognition Models to Occluded Face Recognition

“…The other is the introduction of novel architectures, such as ResNets [13], that pushed the limits of the State-of-the-Art. Finally, an unprecedented proliferation of the internet led to remarkable growth in the data available and how it can be collected [14], with recent works aiming at replacing such data with privacy-friendly synthetic data [15], [16], [17].…”