Vision Transformer-Based Video Hashing Retrieval for Tracing the Source of Fake Videos

<div>Experimental results on three different types of datasets are presented: DeepFake, video inpainting, and video splicing. (a) FF++, (b) DAVIS-VI, and (c) video splicing.</div>

Security and Communication Networks

fig6

Figure 6

Figure 6: Vision Transformer-Based Video Hashing Retrieval for Tracing the Source of Fake Videos 