Training Strategies and Data Augmentations in CNN-based DeepFake Video Detection
This work addresses the problem of unreliable deepfake video detection for users on social media and the internet, but it is incremental as it focuses on optimizing existing methods rather than introducing new paradigms.
The paper analyzes how different training strategies and data augmentation techniques affect CNN-based deepfake detectors, finding that performance varies significantly when training and testing on the same dataset versus across different datasets, with accuracy often limited and biased.
The fast and continuous growth in number and quality of deepfake videos calls for the development of reliable detection systems capable of automatically warning users on social media and on the Internet about the potential untruthfulness of such contents. While algorithms, software, and smartphone apps are getting better every day in generating manipulated videos and swapping faces, the accuracy of automated systems for face forgery detection in videos is still quite limited and generally biased toward the dataset used to design and train a specific detection system. In this paper we analyze how different training strategies and data augmentation techniques affect CNN-based deepfake detectors when training and testing on the same dataset or across different datasets.