English

SpectraNet: FFT-assisted Deep Learning Classifier for Deepfake Face Detection

Computer Vision and Pattern Recognition 2025-11-25 v1 Machine Learning

Abstract

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.

Keywords

Cite

@article{arxiv.2511.19187,
  title  = {SpectraNet: FFT-assisted Deep Learning Classifier for Deepfake Face Detection},
  author = {Nithira Jayarathne and Naveen Basnayake and Keshawa Jayasundara and Pasindu Dodampegama and Praveen Wijesinghe and Hirushika Pelagewatta and Kavishka Abeywardana and Sandushan Ranaweera and Chamira Edussooriya},
  journal= {arXiv preprint arXiv:2511.19187},
  year   = {2025}
}

Comments

4 pages, 3 figures

R2 v1 2026-07-01T07:52:16.804Z