[AISWorld] FSU Develops Veracity AI for deepfake image manipulation detection

[Shuyuan Metcalfe]

Dear colleagues,



We have a new research artifact—called Veracity AI—to be shared as a tool
for detecting deepfake imagery. This tool is focused on analyzing
photographic images that have been manipulated (not for AI-generated
deepfake images). Our fully functional image manipulation platform is now
available to be used for classes, research, or image verification. You are
invited to give it a try.



Here is the URL to the tool: https://veracity.cci.fsu.edu/. Click on “Try
Me!”



The prediction architecture uses a deep consensus algorithm running on an
on-site set of high-performance computing (HPC) nodes, hosted at Florida
State University Research Computing Center. The deep consensus model
generates a prediction mask, indicating where manipulations have occurred
on a pixel level.



You’re encouraged to use this as a tool for yourselves, or as an
instructional material in the classroom when you discuss the problem of
disinformation or deepfake in society. As this is a very new technology,
your feedback is welcome!



Please contact me with research collaboration opportunities. I am open to
ideas that explore the problems of deepfake imagery.



Looking forward to hearing from you!



Thanks,

Shuyuan



--
Shuyuan Metcalfe | Associate Professor
Florida State University School of Information

Trusted CI Fellow, The NSF Cybersecurity Center for Excellence
President-Elect, AIS SIG Information Security & Privacy
Track Co-Chair, AIS SIGSEC Information Security & Privacy Track at 2023
AMCIS, Panama City, Panama