On the Study of Biometric Spoofing Detection using Deep Learning

2026-06-09

AI Summary

This study addresses the vulnerability of face anti-spoofing systems to presentation attacks by systematically evaluating the performance of MobileNetV2, DenseNet-121, Inception-v3, and Spoof Trace Disentanglement (STD) on the CelebA-Spoof and MSU-MFSD datasets, along with an analysis of their cross-dataset generalization capabilities. Experimental results demonstrate that MobileNetV2 achieves the best performance with 92% accuracy, striking an effective balance between efficiency and precision, which renders it well-suited for real-world deployment. Inception-v3 exhibits moderate robustness, whereas DenseNet-121 and STD show limited generalization across datasets. The findings underscore the practical advantages of lightweight architectures and offer insights for future research in domain adaptation and hybrid model design.

Abstract

Biometric systems are increasingly deployed in security applications; however, they remain vulnerable to spoofing attacks, in which attackers exploit counterfeit biometric data to gain unauthorized access. This research evaluates the effectiveness of state-of-the-art machine learning models, MobileNetV2, DenseNet-121, Inception-v3, and Spoof Trace Disentanglement (STD) in detecting spoofing attacks within facial recognition systems. Using the CelebA-Spoof dataset, the study evaluates model effectiveness using metrics such as accuracy, precision, recall, and F1 Score. Cross-dataset validation is carried out on the MSU-MFSD dataset to assess generalizability. The results show MobileNetV2 as the most efficient model, achieving 92% accuracy while balancing computational effectiveness, making it appropriate for real-life applications. Inception-v3 shows moderate robustness, while DenseNet-121 and STD struggle with generalization. The findings highlight the need for advances in domain adaptation and hybrid architectures to enhance biometric security systems.

Problem

• biometric spoofing
• spoofing detection
• facial recognition
• deep learning
• security

Innovation

• spoofing detection
• deep learning
• cross-dataset validation
• domain adaptation
• biometric security