Since the encoding is performed by a Variational Autoencoder, the VAE encoder must output some mean and log variance that we can use to sample a latent vector of shape (latent_dim,). But then how does Stable Diffusion use a vector of shape 64x64x3?
Reference (https://arxiv.org/pdf/2112.10752.pdf, page 24, table 12):
Since the encoding is performed by a Variational Autoencoder, the VAE encoder must output some mean and log variance that we can use to sample a latent vector of shape (latent_dim,). But then how does Stable Diffusion use a vector of shape 64x64x3?
Reference:
Since the encoding is performed by a Variational Autoencoder, the VAE encoder must output some mean and log variance that we can use to sample a latent vector of shape (latent_dim,). But then how does Stable Diffusion use a vector of shape 64x64x3?
Reference (https://arxiv.org/pdf/2112.10752.pdf, page 24, table 12):
I am wondering why the dimensionality of Latent Space in Stable Diffusion is 64x64x3. Since the encoding is performed by Variational Autoencoder ita Variational Autoencoder, the VAE encoder must output some mean and log variance that we can use to sample a latent vectorlatent vector of shape (latent_dim, ). But then how Stable Diffusion uses 64x64x3 latent space when we expectdoes Stable Diffusion use a vector of shape (latent, )64x64x3?
Reference:
I am wondering why the dimensionality of Latent Space in Stable Diffusion is 64x64x3. Since the encoding is performed by Variational Autoencoder it must output some mean and log variance that we can use to sample a latent vector of shape (latent_dim, ). But how Stable Diffusion uses 64x64x3 latent space when we expect shape (latent, )?
Reference:
Since the encoding is performed by a Variational Autoencoder, the VAE encoder must output some mean and log variance that we can use to sample a latent vector of shape (latent_dim,). But then how does Stable Diffusion use a vector of shape 64x64x3?
Reference:
I am wondering why the dimensionality of Latent Space in Stable Diffusion is 64x64x3. Since the encoding is performed by Variational Autoencoder it must output some mean and log variance that we can use to sample a latent vector of shape (latent_dim, ). But how Stable Diffusion uses 64x64x3 latent space when we expect shape (latent, )?
Reference:
I am wondering why the dimensionality of Latent Space in Stable Diffusion is 64x64x3. Since the encoding is performed by Variational Autoencoder it must output some mean and log variance that we can use to sample a latent vector of shape (latent_dim, ). But how Stable Diffusion uses 64x64x3 latent space when we expect shape (latent, )?
I am wondering why the dimensionality of Latent Space in Stable Diffusion is 64x64x3. Since the encoding is performed by Variational Autoencoder it must output some mean and log variance that we can use to sample a latent vector of shape (latent_dim, ). But how Stable Diffusion uses 64x64x3 latent space when we expect shape (latent, )?
Reference:
