Skip to content

Foundation Models for Face Presentation Attack Detection: A Unified Linear-Probing Benchmark

Jul 2026 · arXiv.org · Vol abs/2607.26993 · 1 citation · 67 references
Computer Science

TL;DR

It is suggested that while pretrained representations contain PAD-relevant information, explicit adaptation remains necessary to address domain shift.

Abstract

Face presentation attack detection (PAD) remains challenging under cross-dataset evaluation, where domain shift degrades models trained on a single dataset. The scarcity of large-scale labeled data motivates adapting pretrained vision models rather than training task-specific architectures from scratch, raising a fundamental question: do general-purpose vision foundation models encode PAD-relevant information accessible with minimal task-specific training? To investigate, we systematically evaluate 24 frozen encoders, including self-supervised vision transformers, vision-language encoders, and supervised CNNs, using a unified linear-probing protocol on the MCIO benchmark (MSU-MFSD, CASIA-FASD, Replay-Attack, OULU-NPU). The backbone remains fixed, and only a lightweight linear head is trained to isolate the PAD information already present in the pretrained representation. Results show that frozen foundation-model representations can support strong intra-dataset PAD performance with only a linear classifier, but this performance does not reliably transfer across datasets. Model scale is beneficial within several families, although the effect is not monotonic and is strongly mediated by architecture and pretraining. InternViT-6B achieves the lowest mean intra-dataset error, whereas CLIP ViT-B/32 offers the most favorable cross-dataset transfer-compute trade-off among the evaluated probes. These findings suggest that while pretrained representations contain PAD-relevant information, explicit adaptation remains necessary to address domain shift.

View source

Similar papers

Preprint Aug 2026

Foundation and Multimodal Large Language Models for Face Presentation and Morph Attack Detection

The experiments show that FMs and MLLMs can achieve significant performance for PAD and MAD and the fine-tuned models achieve state-of-the-art detection performance in cross-dataset evaluation, indicating that general-purpose pretrained representations carry substantial attack-relevant information.

H. O. Shahreza, A. H. Khan, P. Lorenz et al. · 0 citations
Preprint Aug 2026

LoRA-based Adaptation Alone Is Not Enough: Understanding the Limits of Foundation Models for Face Presentation Attack Detection

Face presentation attack detection (PAD) aims to reliably detect a wide range of presentation attacks. While PAD methods achieve strong performance within individual datasets, their performance degrades under cross-dataset evaluation. Variations in sensors or lighting conditions can reduce the effectiveness of detector...

P. Lorenz, Anjith George, Marcel Sébastien · 0 citations
Conference Open access Aug 2026

Deep fake detection in face images and video by tuning a pretrained CLIP model

This study develops a video deepfake detection system that addresses the critical challenge of cross-dataset generalization in real-world scenarios. It adopts CLIP (Contrastive Language-Image Pre-training) as the foundation, leveraging its strong vision-language representations for detecting subtle facial manipulations...

Mao-Mao Ling, Włodzimierz Kasprzak · 0 citations
#artificial intelligence Preprint Sep 2026

Swin Meets EfficientNet: Lightweight Architectures for GAN-Based Face Forensics

This research focuses specifically on GAN-generated synthetic faces, which underpin many face-centric deepfakes, and investigates efficient detection approaches using image analysis alone, suggesting that combining hierarchical CNN features with shifted-window self-attention provides an efficient and computationally li...

S. Basu, Ashima Sood, Vijay Kumar et al. · 0 citations
Preprint Sep 2026

Benchmarking Spatial, Spectral, and Self-Supervised Cues for Face Forgery Detection under Realistic Degradation

The results show that clean-set performance is not a reliable indicator of robustness under compression, resizing, and blurring, and reinforce the need for degraded evaluation protocols and highlight self-supervised visual representations as a promising direction for robust face forgery detection.

L. Cunha, Lucas Sotomaior, Lucas Gasperin et al. · 0 citations
Book Open access Aug 2026

Tomatoes, Potatoes, and Onions: Questioning the Need for Faces in Face Presentation Attack Detection

Examination of TPO, a controlled face-free presentation attack dataset consisting of bona fide, print, and replay recordings of, almost randomly chosen, tomatoes, potatoes, and onions, provides empirical evidence that transferable presentation attack representations can be learned independently of facial content.

Guray Ozgur, Fadi Boutros, N. Damer · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.