Deep Fake Detection Using Convolutional Neural Networks

  • Unique Paper ID: 167417
  • Volume: 11
  • Issue: 3
  • PageNo: 1373-1378
  • Abstract:
  • The rise of DeepFake technology, which utilizes advanced deep learning techniques to create highly convincing but deceptive media, presents substantial challenges to the authenticity of digital content. This research introduces a new methodology for detecting DeepFakes, employing Convolutional Neural Networks (CNNs) for image analysis and a combination of CNNs with Recurrent Neural Networks (RNNs) for video analysis. Our CNN architecture is designed to extract and classify spatial features from images, while the CNN-RNN hybrid model addresses both spatial and temporal dimensions in video data. Through extensive evaluation using metrics such as accuracy, precision, recall, F1-score, and Area Under the ROC Curve (AUC), we demonstrate that our proposed models offer significant improvements in detecting DeepFake content. The results suggest that our approach is effective in distinguishing between genuine and manipulated media, providing a valuable tool for ensuring digital media integrity. This work not only advances detection techniques but also contributes to the broader objective of maintaining trustworthiness in digital communications.

Cite This Article

  • ISSN: 2349-6002
  • Volume: 11
  • Issue: 3
  • PageNo: 1373-1378

Deep Fake Detection Using Convolutional Neural Networks

Related Articles