Deepfake Detection with Machine Learning using CNN and RESNET algorithm

Video forgery is continuously increasing in the digital world due to breaches of information security, consequently establishing a scenario for image and video content monitoring for forgery identification. The spread of fake videos raises security risks and anarchy in society. The reason for video forgery is to the augmentation in the malware, which has facilitated user (anyone) to upload, download, or share objects online comprising audio, images, or video. With the development of technology and ease of creation of fake content, the manipulation of media is carried out on a large scale in recent times. Video forgery detection has applications in media science, forensic analysis, digital investigations, and authenticity verification of a video. The purpose of video forensic technology is to extract features to distinguish fake content frames from original videos. Deepfake media has posed a great threat to media integrity and is being produced and spread widely across social media platforms, the detection of which is seen to be a major challenge.  Proposed, an approach for Deepfake detection has been provided for forgery detection in video. ResNet, a Convolutional Neural Network (CNN) algorithm is used as an approach to detect the Deepfake videos. The model aims to enhance the performance of detecting forgery videos produced by a certain method as well as enhance the accuracy of the detector. The proposed approach only uses the deep features extracted from the ResNet CNN model then applies the conventional mathematical approach on these features to find the forgery in the video. It is the detector to be constantly updated with real-world data, and propose an initial solution in hopes of solving Deepfake video detection.