Real-Time Spatio-Temporal Deepfake Detection for Live Biometric Authentication via EfficientNet-GRU
Deepfake technology poses a critical threat to live video conferencing and biometric authentication. Existing detection models are either purely spatial—rendering them vulnerable to video compression—or rely on computationally heavy 3D-CNNs incompatible with strict real-time CPU latency constraints. We propose a highly...