defense arXiv Nov 24, 2025 · Nov 2025
Nithira Jayarathne, Naveen Basnayake, Keshawa Jayasundara et al. · University of Moratuwa · University of Technology Sydney
Proposes EfficientNet-B6 + FFT hybrid detector for deepfake faces, achieving 91% accuracy with balanced batch training
Output Integrity Attack vision
Detecting deepfake images is crucial in combating misinformation. We present a lightweight, generalizable binary classification model based on EfficientNet-B6, fine-tuned with transformation techniques to address severe class imbalances. By leveraging robust preprocessing, oversampling, and optimization strategies, our model achieves high accuracy, stability, and generalization. While incorporating Fourier transform-based phase and amplitude features showed minimal impact, our proposed framework helps non-experts to effectively identify deepfake images, making significant strides toward accessible and reliable deepfake detection.
cnn University of Moratuwa · University of Technology Sydney
ML Security Papers