Detecting Deep Fakes Video through Media Forensics
Several software tools are available that allow users to easily manipulate the faces and audio in videos. These videos, called deepfakes, are usually harmless, but there is the potential for them to be created to sway political opinion or hurt a person's reputation.
Purdue engineers are working to identify examples of video tampering as part of more extensive research in media forensics. They’ve developed sophisticated machine learning techniques to create an algorithm that detects deepfakes.