0

I am currently reading the ESRGAN paper and I noticed that they have used Relativistic GAN for training discriminator. So, is it because Relativistic GAN leads to better results than WGAN-GP?

SANJAY BHANDARI
  • 11
  • 3
  • Please clarify your specific problem or provide additional details to highlight exactly what you need. As it's currently written, it's hard to tell exactly what you're asking. – Community Oct 12 '22 at 08:53

2 Answers2

1

For anyone looking for answer, as explained in Paper: The relativistic discriminator: a key element missing from standard GAN!,

Yes, Standard RaGAN with gradient penalty generate data of better quality than WGAN-GP while only requiring a single discriminator update per generator update (reducing the time taken for reaching the state-of-the-art by 400%).The images generated by RaGAN are of significantly better quality than the ones generated by WGAN-GP and SGAN with spectral normalization.

SANJAY BHANDARI
  • 11
  • 3
1

There is no one-size-fits-all answer to this question, as the two Generative Adversarial Network (GAN) models – Relativistic GAN and WGAN-GP – each have their own advantages and disadvantages. However, some experts believe that, overall, Relativistic GAN may be slightly better than WGAN-GP, due to its improved stability and performance on a variety of tasks.

One reason why Relativistic GAN may be slightly better than WGAN-GP is that it is generally more stable, meaning it is less likely to produce artifacts or results that are significantly different from the real data. Additionally, Relativistic GAN has shown better results on a variety of tasks, including image generation, object detection, and image super-resolution.

Faizy
  • 1,074
  • 1
  • 6
  • 30