Anand Bhattad, Viraj Shah, Derek Hoiem, D. A. Forsyth
TL;DR我们提出了一种名为make it so的新型GAN倒置方法，可以在噪声空间中操作，保留编辑功能，优于现有方法并实现更高质量的编辑效果。
Abstract
stylegan's disentangled style representation enables powerful image editing by manipulating the latent variables, but accurately mapping real-world images to their latent variables (→