acm-header
Sign In

Communications of the ACM

ACM TechNews

Deepfake Detector Can Spot Real/Fake Video from Blood Flow


View as: Print Mobile App Share:
Assessing the authenticity of a video.

Intel says its FakeCatcher can detect a deepfake in real time by "assessing what makes us human–'blood flow' in the pixels of a video."

Credit: Witthaya Prasongsin/Getty

Semiconductor giant Intel claims its FakeCatcher technology can detect whether videos are genuine or deepfakes with 96% accuracy in real time.

According to the company, the technology evaluates "what makes us human—'blood flow' in the pixels of a video."

Intel explained FakeCatcher can spot color changes in a person’s veins based on how blood circulates throughout the body.

The technology collects signals of blood flow from the face, which algorithms measure to determine the video's authenticity.

From ZDNet
View Full Article

 

Abstracts Copyright © 2021 SmithBucklin, Washington, DC, USA


 

No entries found

Sign In for Full Access
» Forgot Password? » Create an ACM Web Account