Using Benford’s Law for Deepfake Detection

The developments in home computers, united with the thousands upon thousands of images/videos of individuals present on the Internet, allowed for the proliferation of deepfaked media affecting the lives of private individuals and the dangerous spread of misinformation. There’s need for a consistent method of detection of the tampered media as to impede these negative aspects. This work investigates the use of Benford’s Law to detect deepfaked material by analyzing the frequency domain of bonafide facial images and the deepfakes, searching for a significant deviation in their value distribution that may be used to distinguish between the two. Through the approaches presented in this paper, this deviation was not found.