Abstract
Facial recognition has become a widely used method for authentication and identification, with applications for secure access and locating missing persons. Its success is largely attributed to deep learning, which leverages large datasets and effective loss functions to learn discriminative features. Despite these advances, facial recognition still faces challenges in explainability, demographic bias, privacy, and robustness to aging, pose variations, lighting changes, occlusions, and facial expressions. Privacy regulations have also led to the degradation of several datasets, raising legal, ethical, and privacy concerns. Synthetic facial data generation has been proposed as a promising solution. It mitigates privacy issues, enables experimentation with controlled facial attributes, alleviates demographic bias, and provides supplementary data to improve models trained on real data. This study compares the effectiveness of synthetic facial datasets generated using different techniques in facial recognition tasks. We evaluate accuracy, rank-1, rank-5, and the true positive rate at a false positive rate of 0.01% on eight leading datasets, offering a comparative analysis not extensively explored in the literature. Results demonstrate the ability of synthetic data to capture realistic variations while emphasizing the need for further research to close the performance gap with real data. Techniques such as diffusion models, GANs, and 3D models show substantial progress; however, challenges remain.
Abstract (translated)
面部识别已成为认证和身份识别中广泛应用的方法，应用于安全访问和个人寻找等场景。其成功主要归功于深度学习技术，该技术利用大规模数据集和有效的损失函数来学习区分性特征。尽管取得了这些进展，面部识别仍然面临着解释性、人口统计学偏差、隐私以及应对年龄变化、姿势变换、光照改变、遮挡及表情差异等方面的挑战。随着隐私法规的出台，许多数据集也因此遭到破坏或无法使用，引发了法律、伦理和隐私方面的担忧。合成面部数据生成被提出作为解决这些问题的一个有前景的方法。它可以减轻隐私问题，使人们能够在控制的人脸属性条件下进行实验，缓解人口统计学偏差，并提供补充数据以改进基于真实数据训练的模型。本研究比较了采用不同技术生成的合成面部数据集在面部识别任务中的有效性。我们在八个领先的数据集中评估了准确率、第一匹配率（rank-1）、第五匹配率（rank-5）以及在0.01%错误接受率下的真阳性率，提供了文献中未被广泛探讨的一个对比分析。结果表明，合成数据能够捕捉到现实变化的特征，同时也强调了需要进一步研究以缩小与真实数据性能之间的差距。扩散模型、GANs和3D模型等技术已经取得了显著的进步；然而，依然存在挑战有待克服。
URL
https://arxiv.org/abs/2512.05928
