Future - Oriented Deepfake Detection on Human Face: A Systematic Review

Rapid technological advancements and easy access to the web have allowed users and communities to interact with each other on social platforms. Combined with the progress in Generative Artificial Intelligence (GenAI) systems, it has allowed the production of digital content that has a realistic flavor. Due to the advancements in Generative Adversarial Networks (GAN), one can create fake images, audio and video streams of individuals or use their audio and visual information to fit other environments. With recent advancements in deepfake technology, it is possible to generate convincing deepfakes in real-time, therefore, deepfakes are specifically employed to spread fake information and propaganda on social circles that tarnish the reputation of an individual or an organization. Recently, many surveys have focused on generating and detecting deepfake images, audio, and video streams. The study discusses existing deepfake models, detection techniques and the future directions.