3 ms·
The second thing you mention seems largely unrelated to the first. Deepfakes (especially live deepfakes) require extensive learning on extensive data sets. Sure
by jessaustin 8y ago
The second thing you mention seems largely unrelated to the first. Deepfakes (especially live deepfakes) require extensive learning on extensive data sets. Sure, you could do facial recognition that way, but useful facial recognition would learn from a single still photo.