Vision-only models achieve SOTA in face anti-spoofing benchmarks

Researchers have developed a new vision-only baseline for Face Anti-Spoofing (FAS) that demonstrates superior performance and efficiency compared to existing multimodal approaches. The study systematically benchmarks 15 pre-trained vision models, finding that self-supervised models like DINOv2 with Registers are particularly effective at capturing subtle spoofing cues. When combined with specific data augmentation techniques and loss functions, this vision-only approach achieves state-of-the-art results on challenging cross-domain FAS protocols, even under data-constrained conditions. AI

IMPACT Optimizes vision-only models for security applications, potentially reducing computational costs for facial recognition systems.

RANK_REASON Academic paper detailing a new methodology and benchmark results in computer vision. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CV →

• APL
• Attention-weighted Patch Loss
• DINOv2
• Face Anti-Spoofing Data Augmentation
• FAS-Aug
• Limited Source Domains
• MICO
• Mika Feng
• Patch-wise Data Augmentation
• Progressive Democrats of America

• paper
• model release

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