Real vs Fake Detection and Face Recognition

It is now much simpler to develop, change, and create spectacular photos because to enormous breakthroughs in image processing and AI computation. Using the development of current picture editing technologies has made it much simpler to create phone photographs and real images, such as substituting your own face with someone else's. Generic human images can also be produced using generic adversarial networks (GANs). When utilized to create distinguishable fake and real evidence, fake photographs and real images have the ability to mishandle data, hurt individuals, and cause a variety of other issues. In this study, we suggested Fake or real Face Detection and Recognition of, a criminological picture stage employing neural tissue to differentiate between different false face images, and a neural tissue-based classifier to detect fake or real  human appearances. We are concentrating on recognizing and detecting fraudulent photos that are produced both artificially by Generative Adversarial Networks and physically by humans. We also let a trusted opponent who has full access to edit and remove the initial image's metadata whenever they like. We demonstrate the great accuracy of Fake Face Detect in identifying phony face photos produced by both humans and Generative Adversarial Networks. Thus, identifying false facial photos is essential for defending against many forms of exploitation.

Keywords: Mobile Net, LBPH Face recognizer, Real or fake face detection.