通过角度扰动提升生成人脸的多样性，增强识别模型泛化能力
IDperturb: Enhancing Variation in Synthetic Face Generation via Angular Perturbation
- 在单位超球面上对身份嵌入进行角度扰动，生成多样化条件向量
- 使用预训练扩散模型生成视觉多样且身份一致的人脸图像
- 适用于需要高多样性合成数据的面部识别系统训练
合成数据已成为训练面部识别（FR）系统的重要替代方案，尤其在真实生物特征数据因隐私与法律限制难以使用时。近年来，身份条件扩散模型已能生成逼真且身份一致的人脸图像。然而，这些模型普遍存在类内变化不足的问题，影响FR模型的鲁棒性与泛化能力。本文提出IDPERTURB，一种简单有效的几何驱动采样策略，通过在单位超球面的限定角度区域内扰动身份嵌入，生成多样化嵌入向量，无需修改生成模型本身。每个扰动后的嵌入作为预训练扩散模型的条件输入，实现视觉多样且身份一致的人脸图像合成，适用于训练通用性强的FR系统。实验表明，使用IDPERTURB生成的数据训练的FR模型在多个基准测试中表现优于现有合成数据方法。
原文摘要 · Abstract (English)
Synthetic data has emerged as a practical alternative to authentic face datasets for training face recognition (FR) systems, especially as privacy and legal concerns increasingly restrict the use of real biometric data. Recent advances in identity-conditional diffusion models have enabled the generation of photorealistic and identity-consistent face images. However, many of these models suffer from limited intra-class variation, an essential property for training robust and generalizable FR models. In this work, we propose IDPERTURB, a simple yet effective geometric-driven sampling strategy to enhance diversity in synthetic face generation. IDPERTURB perturbs identity embeddings within a constrained angular region of the unit hyper-sphere, producing a diverse set of embeddings without modifying the underlying generative model. Each perturbed embedding serves as a conditioning vector for a pre-trained diffusion model, enabling the synthesis of visually varied yet identity-coherent face images suitable for training generalizable FR systems. Empirical results demonstrate that training FR on datasets generated using IDPERTURB yields improved performance across multiple FR benchmarks, compared to existing synthetic data generation approaches.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。