The paper proposes DBaGNet, a triplet-trained classifier over fused identity, behavioral, and geometric features, reporting strong in-dataset accuracy and cross-dataset AUC gains, though the main cross-dataset result relies on an unspecified augmentation variant.
Automatic face reenactment,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
other 1
citation-polarity summary
fields
cs.CV 1years
2024 1verdicts
REJECT 1roles
other 1polarities
unclear 1representative citing papers
citing papers explorer
-
Securing Social Media Against Deepfakes using Identity, Behavioral, and Geometric Signatures
The paper proposes DBaGNet, a triplet-trained classifier over fused identity, behavioral, and geometric features, reporting strong in-dataset accuracy and cross-dataset AUC gains, though the main cross-dataset result relies on an unspecified augmentation variant.