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.

翻译：我们提出了一种在高质量、多模态数据集上实现快速、逼真图像生成的方法，该方法使用在确定性自编码器潜空间上训练的分层变分自编码器（VAE）。在这一两阶段框架中，自编码器首先将图像压缩为语义特征，随后通过深度VAE对这些特征进行建模。通过这种方式，VAE能够避免对构成图像编码长度主要部分的精细细节进行建模，从而专注于学习图像的结构性成分。我们验证了这一两阶段方法的有效性，在ImageNet-256数据集上达到了9.34的FID分数，与BigGAN性能相当。我们已将实现代码公开提供。