Answers are grounded in this paper and cite their sources.
Investigation shows AI achieved over 95% accuracy in detecting deepfakes in videos, suggesting novel detection methods.
The possibility of deepfake detection using vibraimage technology and pre-trained AI was investigated. The database of 100 deepfake 60-second videos, resolution from 640×480 to 1960×1080 and video compression level from 1 to 10,000 was created. Developed and pre-trained on digital parameters of fast vibraimage transform (FVT) of deepfakes and control (original) video groups, the three-layer ANN showed deepfake detection accuracy over 95% for 10-seconds video processing. The dependences of AI training accuracy of training and testing groups, the dependence of deepfake detection accuracy on the processing time and ROC curves based on specificity and sensitivity data are presented.
No takes yet. Share an insight, caveat, or question.
Акимов et al. (2025) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: