Learning the Manifold of Authenticity: Hybrid-Curvature Representation Learning for Generalizable Deepfake Detection

The practical utility of deepfake detectors is crippled by a crisis of generalization: models that perform well on known manipulation techniques consistently fail when faced with unseen forgeries. We argue this failure stems from a fundamental geometric mismatch. Existing methods implicitly assume t...

Täydet tiedot

Bibliografiset tiedot

Päätekijät:	Nicolas Larue, Vitomir Struc, Peter Peer, Ngoc-Son Vu
Aineistotyyppi:	Artigo
Kieli:	Inglês
Julkaistu:	IEEE 2026-01-01
Sarja:	IEEE Access
Aiheet:	Contrastive learning generalized deepfake detection hyperbolic one-class anomaly detection
Linkit:	https://ieeexplore.ieee.org/document/11557307/
Tagit:	Lisää tagi Ei tageja, Lisää ensimmäinen tagi!