Review on DeepFake Video Detection Techniques

Abstract

Over the last few decades, advanced methods and different tools for manipulating media have been used for good and better results. It is because of increasing progress in artificial intelligence, machine learning, and deep learning methods. Although the technology has typically been utilized for legal reasons, such as in entertainment and education; corrupt people have also taken advantage of it. For instance, realistic and high-quality fake movies, photographs, or audios have been produced to distribute propaganda and false information, incite political unrest and hatred or even frighten and blackmail individuals. A deepfake is a video created using artificial intelligence to show someone saying or doing something they did not do. Since then, many researches have addressed the issues presented and detected by Deepfake videos. We summarized 50 relevant articles from 2018 to 2022 that covered a range of techniques in order to present the advancement of research activities in the field of Deepfake video detection. The main focus of this study is to identify the limitations of previous literature and suggest future direction for upcoming researchers.