thanks for the explanation!
denoising in latent space certainly seems like the "correct" path. My (amateur) thinking is, the more you can do in latent space, the better.
denoising in latent space certainly seems like the "correct" path. My (amateur) thinking is, the more you can do in latent space, the better.
The former seems likely to be lower compute-for-resolution, but that’s not the only consideration for “better”...