SELFI adaptively fuses face identity embeddings with visual features via a learned relevance gate, improving cross-dataset deepfake detection AUC by about 3% on average.
End- to-end reconstruction-classification learning for face forgery detection
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
SELFI: Selective Fusion of Identity for Generalizable Deepfake Detection
SELFI adaptively fuses face identity embeddings with visual features via a learned relevance gate, improving cross-dataset deepfake detection AUC by about 3% on average.