Skip to main content

New answers tagged

0 votes

Why is the variational auto-encoder's output blurred, while GANs output is crisp and has sharp edges?

Fundamentally speaking, the optimization target of a VAE has two parts: minimizing reconstruction loss (similar to an AutoEncoder) and minimizing KL divergence loss (ensuring that the latent z follows ...
hyukshuai lee's user avatar

Top 50 recent answers are included