Abstract
Identical twin face verification represents an extreme fine-grained recognition challenge where even state-of-the-art systems fail due to overwhelming genetic similarity. Current face recognition methods achieve over 99.8% accuracy on standard benchmarks but drop dramatically to 88.9% when distinguishing identical twins, exposing critical vulnerabilities in biometric security systems. The difficulty lies in learning features that capture subtle, non-genetic variations that uniquely identify individuals. We propose the Asymmetric Hierarchical Attention Network (AHAN), a novel architecture specifically designed for this challenge through multi-granularity facial analysis. AHAN introduces a Hierarchical Cross-Attention (HCA) module that performs multi-scale analysis on semantic facial regions, enabling specialized processing at optimal resolutions. We further propose a Facial Asymmetry Attention Module (FAAM) that learns unique biometric signatures by computing cross-attention between left and right facial halves, capturing subtle asymmetric patterns that differ even between twins. To ensure the network learns truly individuating features, we introduce Twin-Aware Pair-Wise Cross-Attention (TA-PWCA), a training-only regularization strategy that uses each subject's own twin as the hardest possible distractor. Extensive experiments on the ND_TWIN dataset demonstrate that AHAN achieves 92.3% twin verification accuracy, representing a 3.4% improvement over state-of-the-art methods.
Abstract (translated)
同卵双胞胎面部验证代表了一个极端的细粒度识别挑战，即使最先进的系统也因为基因相似度过高而无法准确区分。目前的人脸识别方法在标准基准测试中能实现超过99.8%的准确性，但在分辨同卵双胞胎时这一数字会急剧下降至88.9%，这揭示了生物特征安全系统的重大漏洞所在。问题在于如何学习能够捕捉细微非遗传变异特征的能力，这些变异有助于唯一识别个体。 我们提出了一种新颖的架构——不对称分层注意网络（AHAN），特别针对此挑战而设计，通过多粒度面部分析实现功能。AHAN引入了一个层次交叉注意力（HCA）模块，该模块在语义面部区域执行多层次分析，并能在最优分辨率下进行专业化处理。此外，我们还提出了一种面部不对称注意模块（FAAM），它通过计算左右半边脸部之间的跨注意力来学习独特的生物特征签名，捕捉细微的非对称模式，即使对于双胞胎来说也是不同的。 为了确保网络能学习真正具有区分性的特征，我们引入了同卵意识配对交叉注意训练方法（TA-PWCA），这是一种仅用于训练阶段的正则化策略，通过将每个主体与其自己的同卵双胞胎作为最难干扰项来实现。在ND_TWIN数据集上的广泛实验表明，AHAN实现了92.3%的同卵双胞胎验证准确性，相比现有最佳方法提高了3.4%。
URL
https://arxiv.org/abs/2602.21503
