README.md

metadata

base_model:
  - dima806/deepfake_vs_real_image_detection

DeepDefend - Deepfake Analyzer v1

This is a PyTorch model trained to analyze and detect deepfake videos based on cropped face images.

🧠 Model Details

• Architecture: ViTBinary (Vision Transformer)
• Input Size: 224×224
• Framework: PyTorch
• Validation Accuracy: 91.33%
• Trained on: Deepfake Detection Dataset (cropped faces)

📁 Files in this repository

• pytorch_model.bin — Model weights
• config.json — Model configuration details
• README.md — This file

🚀 How to Use

from huggingface_hub import hf_hub_download
import torch
from your_model_file import DeepfakeDetector  # replace with your model class

# Download weights from Hugging Face Hub
model_path = hf_hub_download(repo_id="rushild25/DeepDefend", filename="pytorch_model.bin")

# Initialize model and load weights
model = DeepfakeDetector()
model.load_state_dict(torch.load(model_path, map_location="cpu"))
model.eval()

👤 Author
Uploaded by rushild25