Deep learning models show promise in detecting facial spoofing attacks

This research paper investigates the use of deep learning models, specifically MobileNetV2, DenseNet-121, and Inception-v3, for detecting spoofing attacks in facial recognition systems. Using the CelebA-Spoof dataset, the study found MobileNetV2 to be the most effective, achieving 92% accuracy while maintaining computational efficiency. The paper also highlights the challenges of generalization for other models and suggests future work on domain adaptation and hybrid architectures to improve biometric security. AI

IMPACT Enhances understanding of deep learning's role in securing biometric systems against sophisticated spoofing techniques.

RANK_REASON The cluster contains an academic paper detailing research findings on deep learning models for biometric spoofing detection. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• CelebA-Spoof dataset
• DenseNet-121
• Inception-v3
• MobileNetV2
• MSU-MFSD dataset
• Spoof Trace Disentanglement (STD)

• paper
• safety

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →