5,337
edits
SinGAN: Learning a Generative Model from a Single Natural Image: Difference between revisions
SinGAN: Learning a Generative Model from a Single Natural Image (view source)
Revision as of 18:44, 6 December 2019
, 6 December 2019→Basic Idea
| Line 16: | Line 16: | ||
==Basic Idea== | ==Basic Idea== | ||
<!-- | |||
This is wrong there are no patches in the code. | |||
Bootstrap patches of the original image and build GANs which can add fine details to blurry patches at different path sizes. | Bootstrap patches of the original image and build GANs which can add fine details to blurry patches at different path sizes. | ||
--> | |||
Train GANs to fill in details at different scales of the image | |||
* Start by building a GAN to generate low-resolution versions of the original image | * Start by building a GAN to generate low-resolution versions of the original image | ||
* Then upscale the image and build a GAN to add details to | * Then upscale the image and build a GAN to add details to your upscaled image | ||
* Fix the parameters of the previous GAN. Upscale the outputs and repeat. | * Fix the parameters of the previous GAN. Upscale the outputs and repeat. | ||