BriefGPT.xyz
Mar, 2021
StyleGAN循环伪影的系统分析与去除
Systematic Analysis and Removal of Circular Artifacts for StyleGAN
HTML
PDF
Way Tan, Bihan Wen, Xulei Yang
TL;DR
通过对Vanilla StyleGAN模型不同阶段的功能进行机制分析和大量实验调研，我们提出了一种有效的简单解决方案，采用新的像素-实例归一化（PIN）层来消除Vanilla StyleGAN生成图像中引起视觉质量降低的圆形伪影。
Abstract
stylegan
is one of the state-of-the-art image generators which is well-known for synthesizing high-resolution and
hyper-realistic face images
. Though images generated by vanilla
→