Video Face Manipulation Detection Through Ensemble of CNNs
-
Updated
Jun 27, 2024 - Jupyter Notebook
Video Face Manipulation Detection Through Ensemble of CNNs
Code for Video Deepfake Detection model from "Combining EfficientNet and Vision Transformers for Video Deepfake Detection" presented at ICIAP 2021.
A state-of-the-art, open-source deepfake detection system built with PyTorch and EfficientNet-B0, featuring a user-friendly web interface for real-time image and video analysis.
Implementation of the winning solution for the Media Analytics Challenge 2023.
A sleek PyTorch‑powered DeepFake detection pipeline that fuses the state‑of‑the‑art ResNeXt‑50 CNN backbone with an LSTM‑based temporal encoder. Rigorous evaluation on DFDC and AvLips datasets yields 95.19% and 98.88% test accuracy, respectively.
✦ Real-Time AI-Powered Deepfake Detection System ✦ Detect Manipulated Images, Videos & Cloned Voices — Powered by Deep Learning
Detect AI-generated face manipulations in real time — with visual explanations showing exactly where the model found tampering. Supports images, videos, and live webcam. Powered by EfficientNet-B4 and Grad-CAM XAI.
Detect deepfake images and videos with EfficientNet-B0, face extraction, Grad-CAM heatmaps, and a FastAPI REST API for clear analysis
A sleek PyTorch-powered DeepFake detection pipeline that combines a ResNeXt-50 CNN backbone with an LSTM-based temporal encoder for robust video forgery detection. Evaluated on the DFDC and AvLips datasets, the model achieves test accuracies of 95.19% and 98.88%, respectively.
To associate your repository with the dfdc topic, visit your repo's landing page and select "manage topics."