Open potentialming opened 3 weeks ago
I have a naive idea. Here you use skeletons as guides. Can the training use canny images or depth maps as guides so that the model can generate videos through canny or depth maps?
I have a naive idea. Here you use skeletons as guides. Can the training use canny images or depth maps as guides so that the model can generate videos through canny or depth maps?