View Complete Reference

Vishnu, U (2021)

Deepfake Detection using Benford’s Law and Distribution Variance Statistic

International Research Journal of Engineering and Technology(IRJET) 08(10), pp. 712-719.

ISSN/ISBN: Not available at this time. DOI: Not available at this time.



Abstract: Deepfake images have been a trending technology for a long time. Though it seems amusing to our eyes how a computer can generate realistic fake images, it has become double trouble for celebrities and Important personnel wherein their faces are morphed to bring down their reputation. Deepfakes can also be used to change the content they speak by morphing lip movements and integrating the video with artificially generated audio. Previous experimentations and research in this field have contributed many methods for detection deepfake media, which unfortunately tends to get fooled by the realism of the artificially generated media. Hence, there is an immediate requirement for a more stable method for deepfake detection. In this paper, we will observe how Benford’s law or Law of the First digit can be used to detect deepfake media.


Bibtex:
@article{, author = {Vishnu, U}, title = {Deepfake Detection using Benford’s Law and Distribution Variance Statistic}, year = {2021}, journal = {International Research Journal of Engineering and Technology(IRJET)}, volume = {8}, issue = {10}, pages = {712--719}, } }


Reference Type: Journal Article

Subject Area(s): Computer Science, Image Processing