Open-source deepfake detection: train your own model with TensorFlow, Keras & EfficientNet
-
Updated
Mar 19, 2026 - Python
Open-source deepfake detection: train your own model with TensorFlow, Keras & EfficientNet
Attempts to bypass AI Image detection by employing various methods such as: Noise Injection, FFT Smoothing, FFT Matching, Pixel Perturbation, Camera Simulation, and more.
This repository employs TensorFlow-Keras and MTCNN (Multi-Task Cascaded Convolutional Networks) to build an efficient deepfake detection system.
Add a description, image, and links to the df-detect topic page so that developers can more easily learn about it.
To associate your repository with the df-detect topic, visit your repo's landing page and select "manage topics."