Research on Active Defense of Facial Images in Deepfakes

Silesian College, Yanshan University, Qinhuangdao, Hebei Province, China

This email address is being protected from spambots. You need JavaScript enabled to view it.

Abstract

In today's society, with the rapid development of generative artificial intelligence, generative adversarial network (GAN) based on deep learning, image synthesis and replacement technology of deepfake technology has reached an unprecedented degree of realism, through accurate learning of face distribution features, deepfake technology can be applied in various software, but everything must be seen from a dialectical point of view, and the attitude of dividing into two makes us pay attention to preventing the harm caused by deepfake technology after being illegally used. Researchers must attach importance to and vigorously develop active defense technology. In today's cognition, active defense is to be more active and defensive advantages than in the detection of passive defense, into the disturbance imaging and watermark information, or for illegal use to trace the source, so this article will state and introduce the existing active defense technology, including active interference and active forensics, from the principle, applicable scenarios and technical algorithms to introduce in detail the advantages and technical cores, Prospects and areas for improvement in order to better counter and defeat the malicious use of deepfakes.