TileGAN：大规模非均质纹理的合成

Apr, 2019

TileGAN: Synthesis of Large-Scale Non-Homogeneous Textures

Anna Frühstück, Ibraheem Alhashim, Peter Wonka

TL;DR在输入多张图像且需要大规模输出的纹理合成问题中，本文基于生成对抗网络提出两个扩展解决方案，一是结合较小分辨率的GAN训练产生边界无明显痕迹的大规模纹理地图，二是提供用户界面进行艺术控制，我们的定量和定性结果展示了合成高分辨率地图的成功案例。

Abstract

We tackle the problem of texture synthesis in the setting where many input images are given and a large-scale output is required. We build on recent generative adversarial networks and propose two extensions in t