Hugging Face's logo

shivani1511
/
deepfake-image-detector-new-latest-v2

Image Classification
Safetensors
vit
deepfake-detection
real-and-fake-faces
Model card Files
xet
Community
deepfake-image-detector-new-latest-v2 / README.md
shivani1511's picture
shivani1511
Fine-tuned model on Real and Fake Faces dataset with improvements
0166304 verified
preview code
|
raw
history blame contribute delete
1.16 kB
metadata
license: mit
tags:
  - deepfake-detection
  - image-classification
  - real-and-fake-faces

Deepfake Image Detector (Real and Fake Faces Fine-Tuned)

Model Performance

  • Test Accuracy: 93.02%
  • Best Validation Accuracy: 94.12%
  • Best Epoch: 1
  • Planned Epochs: 12
  • Actual Epochs Trained: 5 (early stopping applied)

Dataset

  • Training: 1,367 images
  • Validation: 170 images
  • Test: 172 images

Training Details

Training was stopped early at epoch 5 due to early stopping criteria being met. The best model was achieved at epoch 1 with validation accuracy of 94.12%.

Usage 

from transformers import ViTForImageClassification, ViTFeatureExtractor

model = ViTForImageClassification.from_pretrained('shivani1511/deepfake-image-detector-new-latest-v2')
feature_extractor = ViTFeatureExtractor.from_pretrained('shivani1511/deepfake-image-detector-new-latest-v2')

Notes

  • Fine-tuned on Real and Fake Faces dataset to address AI-generated fake detection.
  • Base model: shivani1511/deepfake-image-detector-new-latest (Vision Transformer).
  • Improvements: Enhanced data augmentation, class-weighted loss, Mixup, more unfrozen layers.