High Fidelity Image Synthesis With Deep VAEs In Latent Space
We present fast, realistic image generation on high-resolution, multimodal datasets using hierarchical variational autoencoders (VAEs) trained on a deterministic autoencoder's latent space. In this two-stage setup, the autoencoder compresses the image into its semantic features, which are then modeled with a deep VAE. With this method, the VAE avoids modeling the fine-grained details that constitute the majority of the image's code length, allowing it to focus on learning its structural components. We demonstrate the effectiveness of our two-stage approach, achieving a FID of 9.34 on the ImageNet-256 dataset which is comparable to BigGAN. We make our implementation available online.
PDF Abstract
Datasets
ImageNet
Results from the Paper
Submit
results from this paper
to get state-of-the-art GitHub badges and help the
community compare results to other papers.
Methods
1x1 Convolution •
Adam •
AutoEncoder •
Batch Normalization •
BigGAN •
Conditional Batch Normalization •
Convolution •
Dense Connections •
Dot-Product Attention •
Early Stopping •
Feedforward Network •
GAN Hinge Loss •
Linear Layer •
Non-Local Block •
Non-Local Operation •
Off-Diagonal Orthogonal Regularization •
Projection Discriminator •
ReLU •
Residual Block •
Residual Connection •
SAGAN •
SAGAN Self-Attention Module •
Softmax •
Spectral Normalization •
Truncation Trick •
TTUR •
VAE
