Classifying Real vs. Deepfake Videos using EfficientNet-B0
This project detects whether a video is real or a deepfake using the EfficientNet-B0 architecture. We applied a face-focused preprocessing pipeline to improve detection, achieving 59% validation accuracy.
Run our model directly in Google Colab: Open in Colab .
View the full source code on GitHub: DeepFake Detector Repository .