Hierarchical Quantized Autoencoders
arXiv:2002.08111
Abstract
Despite progress in training neural networks for lossy image compression, current approaches fail to maintain both perceptual quality and abstract features at very low bitrates. Encouraged by recent success in learning discrete representations with Vector Quantized Variational Autoencoders (VQ-VAEs), we motivate the use of a hierarchy of VQ-VAEs to attain high factors of compression. We show that the combination of stochastic quantization and hierarchical latent structure aids likelihood-based image compression. This leads us to introduce a novel objective for training hierarchical VQ-VAEs. Our resulting scheme produces a Markovian series of latent variables that reconstruct images of high-perceptual quality which retain semantically meaningful features. We provide qualitative and quantitative evaluations on the CelebA and MNIST datasets.
References in corpus (27)
- Auto-Encoding Variational Bayes
- GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibrium
- Categorical Reparameterization with Gumbel-Softmax
- Neural Discrete Representation Learning
- Conditional Image Generation with PixelCNN Decoders
- Variational image compression with a scale hyperprior
- The Concrete Distribution: A Continuous Relaxation of Discrete Random Variables
- Toward Multimodal Image-to-Image Translation
- The Perception-Distortion Tradeoff
- Deep Unsupervised Clustering with Gaussian Mixture Variational Autoencoders
- Image and Video Compression with Neural Networks: A Review
- Recent Advances in Autoencoder-Based Representation Learning
- Soft-to-Hard Vector Quantization for End-to-End Learning Compressible Representations
- Neural Variational Inference and Learning in Belief Networks
- Classification Accuracy Score for Conditional Generative Models
- Towards Deeper Understanding of Variational Autoencoding Models
- BIVA: A Very Deep Hierarchy of Latent Variables for Generative Modeling
- A Fast Incremental Gaussian Mixture Model
- Taming VAEs
- Hierarchical Autoregressive Image Models with Auxiliary Decoders
- Computationally Efficient Neural Image Compression
- Deep Perceptual Compression
- Variational Memory Addressing in Generative Models
- Perceptual Generative Autoencoders
- Adaptive Density Estimation for Generative Models
- Hierarchical compositional feature learning
- Neural Image Compression and Explanation