A dual-stream deepfake detector using global context attention and optical flow residuals reports high accuracy on FF++ and Celeb-DF, but with no error bars or code and only marginal gains over prior work.
Beyond the prior forgery knowledge: Mining critical clues for general 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
-
GC-ConsFlow: Leveraging Optical Flow Residuals and Global Context for Robust Deepfake Detection
A dual-stream deepfake detector using global context attention and optical flow residuals reports high accuracy on FF++ and Celeb-DF, but with no error bars or code and only marginal gains over prior work.