stylegan-human / StyleGAN-Human

StyleGAN-Human: A Data-Centric Odyssey of Human Generation
1.15k stars 144 forks source link

You have trained this, the final result is this, is the input a, b images, put in shallow support space to generate x model (a, b corresponds to seed2, 3) then 2, 3 is the original image a, b well, or is his mapping, is the input and generated output is the same image well, one can not be edited, one can be pti inversion gan model #49

Open xuboot opened 1 year ago

xuboot commented 1 year ago

You have trained this, the final result is this, is the input a, b images, put in shallow support space to generate x model (a, b corresponds to seed2, 3) then 2, 3 is the original image a, b well, or is his mapping, is the input and generated output is the same image well, one can not be edited, one can be pti inversion gan model