Understanding SRGANs
SRGANs are a class of GANs that focuses on creating super-resolution images from low-resolution images.
The functionality of an SRGAN algorithm is described as such: the algorithm picks a high-resolution image from the dataset and samples it down to a low-resolution image. Then, the generator neural network tries to produce a higher resolution image from the low-resolution image. We will call this a super-resolution image from now on. The super-resolution image is sent to the discriminator neural network, which has already been trained on samples of high-resolution images and some basic super-resolution images so that they can be classified.
The discriminator classifies the super-resolution image sent to it by the generator ...
Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.
Read now
Unlock full access