AHAN: Asymmetric Hierarchical Attention Network for Identical Twin Face Verification

TL;DR

This paper introduces AHAN, a novel neural network architecture with multi-scale and asymmetric attention modules, significantly improving the accuracy of identical twin face verification by capturing subtle individual differences.

Contribution

The paper proposes the Asymmetric Hierarchical Attention Network (AHAN) with innovative modules for multi-granularity analysis and asymmetric feature learning, addressing the challenge of twin face verification.

Findings

Achieves 92.3% verification accuracy on ND_TWIN dataset.

Improves over state-of-the-art by 3.4%.

Effectively captures subtle biometric differences between twins.

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.