A Novel Method of Deepfake Detection

Deep-Fake is a novel artificial media technology that uses the likeness of someone else to replace people in existing photographs and films. Deep Learning, as name implies, type Artificial Intelligence used create it. It critical develop counter attacking approaches for detecting fraudulent data. This research examines depth. The Detection discussed here based on current datasets, such Challenge (DFDC) Google’s dataset (DFD). creation bespoke from high-quality Deep-Fakes was utilised test models. results both with without Transfer Learning were analysed. Finally, trained models spot well-known deep-fakes former US President Barack Obama actor Tom Cruise. A comparison study performed all three findings show detection are generally domain-specific tasks, however using considerably improves model performance parameters, whereas convolutional RNN gives sequence advantage.