A spatio-temporal attention network with texture enhancement reports 90.91% ACC and 0.9810 AUC on FF++(LQ) and 95.97% ACC on a cross-dataset DFDC test, but the evaluation protocol is incomplete.
Xception: Deep learning with depthwise separable convolutions
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
Deepfake Detection with Spatio-Temporal Consistency and Attention
A spatio-temporal attention network with texture enhancement reports 90.91% ACC and 0.9810 AUC on FF++(LQ) and 95.97% ACC on a cross-dataset DFDC test, but the evaluation protocol is incomplete.