Deep Fake Detection using Deep Learning
An extensive survey on deepfake generation and recognition techniques using neural networks is provided and a detailed study of the different technologies used in deepfake detection is provided.
Abstract
Deep learning is an effective method that is broadly used across a wide range of areas, i.e., computer vision, machine vision, and natural language processing. Deepfakes is an application of this technology where the images and videos of someone are manipulated in such a way that it is difficult for human beings to tell the difference between them and their true selves. Deepfakes have been the subject of several studies recently, and a number of deep learning approaches have been proposed for their detection. Here, we provide an extensive survey on deepfake generation and recognition techniques using neural networks. Additionally, a detailed study of the different technologies used in deepfake detection is provided. This will surely benefit researchers in this area because it will include new cutting-edge methods for detecting fake videos or images on social networks. Moreover, it will make it easy for us to compare what others have done in their papers by explaining how they came up with their models or what information was employed for training them. Key Words: Deep Learning, Fake Detection, Neural Networks, Social Networks