Abstract:
As DEEPFAKE and other artificial intelligence methods have become more and more powerful in tampering with digital multimedia, once again, governments and people around the world are concerned about the issue of digital media tampering. Both traditional tampering attack and intelligent tampering have raised people's doubts about the authenticity of digital media. Therefore, passive detection of digital video tampering traces is of great significance. This paper gives an overview of this problem to analyze the current development status and future trends in this field. First, the background and negative impacts of video tampering are analyzed. The meaning of video tampering detection from different angles, and the difficulties of the problem are also summarized. Next, this paper analyzes different types of digital video re-editing technologies, and divides the existing passive forensic detection technologies into detection method based on spatio-temporal domain feature, coding domain feature, and depth feature, features and deficiencies are detailed in this paper. In addition, this paper also summarizes the existing international and domestic competitions, video tampering data sets, and third-party tool libraries commonly used in the field, and also points out the follow-up research content and development trends in this field.