用于可控人脸合成的自适应对抗增强

中文摘要 AI 辅助

合成数据为训练人脸识别模型提供了真实世界数据集的可扩展替代方案，尤其适用于低分辨率、遮挡和佩戴口罩等具有挑战性的场景。然而，大多数方法缺乏多样性且无法有效泛化。我们提出了Ensemble Feedback Controllable Synthesis（EFCS，集成反馈可控合成），这是一种引导式框架，可生成多样化且具有挑战性的样本，同时保持视觉真实性。EFCS扩大了分布变异性，与单反馈合成和随机合成相比，其FID和KID分数通常更高，同时保持了高精确度。在EFCS数据上训练的识别模型在多个基准测试中始终优于基线，显示出对真实世界场景的泛化能力提升。此外，我们引入了一种基于分析的公式，将扰动诱导的难度、样本效用和性能下降关联起来，为平衡合成数据的复杂度以实现最优训练提供了原则性见解。这些贡献共同确立了EFCS作为一种有效且具有分析基础的方法，可弥合合成数据集与真实数据集之间的差距。

英文摘要

Synthetic data provides a scalable alternative to real-world datasets for training face recognition models, particularly under challenging conditions such as low resolution, occlusion, and masks. Yet, most approaches lack diversity and fail to generalize effectively. We propose Ensemble Feedback Controllable Synthesis (EFCS), a guided framework that generates diverse and challenging samples while preserving visual realism. EFCS expands distributional variability, often reflected in higher FID and KID scores compared to single-feedback and random synthesis, while maintaining high precision. Recognition models trained on EFCS data consistently outperform baselines across multiple benchmarks, showing improved generalization to real-world scenarios. Furthermore, we introduce an analytically motivated formulation linking perturbation-induced difficulty, sample utility, and performance degradation, offering principled insights into balancing synthetic data complexity for optimal training. Together, these contributions establish EFCS as an effective and analytically grounded approach for bridging the gap between synthetic and real datasets.

发表机构

• Manipal University Jaipur（斋浦尔马尼帕尔大学）
• Birla Institute of Technology and Science Pilani Dubai Campus（Pilani贝拉理工学院迪拜校区）
• Indian Institute of Technology Jodhpur（焦特布尔印度理工学院）

机构由 AI 辅助整理，请以论文原文为准。