Image Feature Detectors for Deepfake Image Detection Using Transfer Learning
Khattab M. Ali Alheeti, Salah Sleibi Al-Rawi, Haitham Abbas Khalaf, Duaa Al Dosary · 2021 14th International Conference on Developments in eSystems Engineering (DeSE) · 2021
Deepfake is heavily based on artificial intelligence and machine learning to generate audio content and visuals with an extremely high potential to deceive. In this paper, a new detection system to identify any deepfake for audio or images. In other words, transfer learning techniques are used to present an intelligent detection system of deepfake videos. In this work, a support vector machine is employed for the detection process. Outstanding results are obtained from the train and test this system with various types of images that are real or fake.