domain shift 1face anti-spoofing 1foundation models 1linear probing 1presentation attack detection 1
From the 1 of 2 linked papers with an AI index.
2 papers
cs.CV2026
LoRA-based Adaptation Alone Is Not Enough: Understanding the Limits of Foundation Models for Face Presentation Attack Detection
Peter Lorenz, Anjith George, Marcel Sébastien
Face presentation attack detection (PAD) aims to reliably detect a wide range of presentation attacks. While PAD methods achieve strong performance within individual datasets, thei…
cs.LG2026
Foundation Models for Face Presentation Attack Detection: A Unified Linear-Probing Benchmark
Peter Lorenz, Anjith George, Sébastien Marcel
The paper benchmarks frozen vision foundation models for face presentation attack detection using a linear-probing protocol, finding strong intra-dataset performance but limited cr…