Preprint
Sep 2026
Preserving Knowledge across Space and Time for Continual Video Deepfake Detection
A continual deepfake video detection framework that explicitly decomposes video features into spatial, temporal, and spatiotemporal modalities in the frequency domain, and adopts a cross-modality decorrelation loss that encourages spatiotemporal representations to remain orthogonal to single-modality cues.
Taehoon Kim, Jongwook Choi, Heejae Jo et al.
· 0 citations