China has unveiled a neuron that creates the most lifelike videos possible from a single photo. OmniHuman-1 not only makes the people in the photos speak by precisely moving their lips, but also animates them in a meaningful way based on the context of their surroundings.