Deepfake techniques, which present realistic AI-generated videos of people doing and saying fictional things, have the potential to have a significant impact on how people determine the legitimacy of information presented online. These content generation and modification technologies may affect the quality of public discourse and the safeguarding of human rights—especially given that deepfakes may be used maliciously as a source of misinformation, manipulation, harassment, and persuasion. Identifying manipulated media is a technically demanding and rapidly evolving challenge that requires collaborations across the entire tech industry and beyond.
-
FSGAN: Subject Agnostic Face Swapping and Reenactment[paper]
-
Deepfake Video Detection through Optical Flow based CNN[paper]
-
Celeb-DF: A New Dataset for DeepFake Forensics[paper]
-
FaceForensics++: Learning to Detect Manipulated Facial Images[paper benchmarks github]
-
Multi-task Learning For Detecting and Segmenting Manipulated Facial Images and Videos:[paper]
-
FakeCatcher: Detection of Synthetic Portrait Videos using Biological Signals[paper]
-
Exposing DeepFake Videos By Detecting Face Warping Artifacts[paper]
-
MesoNet: a Compact Facial Video Forgery Detection Network[paper]
-
EXPOSING DEEP FAKES USING INCONSISTENT HEAD POSES[paper]
-
CAPSULE-FORENSICS: USING CAPSULE NETWORKS TO DETECT FORGED IMAGES AND VIDEOS[paper]
-
Detection of Deepfake Video Manipulation[paper]
-
Deepfake Video Detection Using Recurrent Neural Networks[paper]