Hierarchical Autoregressive Modeling for Neural Video Compression
arXiv:2010.10258
Abstract
Recent work by Marino et al. (2020) showed improved performance in sequential density estimation by combining masked autoregressive flows with hierarchical latent variable models. We draw a connection between such autoregressive generative models and the task of lossy video compression. Specifically, we view recent neural video compression methods (Lu et al., 2019; Yang et al., 2020b; Agustssonet al., 2020) as instances of a generalized stochastic temporal autoregressive transform, and propose avenues for enhancement based on this insight. Comprehensive evaluations on large-scale video data show improved rate-distortion performance over both state-of-the-art neural and conventional video compression methods.
arXiv commentPublished as a conference paper at ICLR 2021
References in corpus (14)
- Estimating or Propagating Gradients Through Stochastic Neurons for Conditional Computation
- NICE: Non-linear Independent Components Estimation
- Variational Inference with Normalizing Flows
- Density estimation using Real NVP
- A Recurrent Latent Variable Model for Sequential Data
- Joint Autoregressive and Hierarchical Priors for Learned Image Compression
- Stochastic Video Generation with a Learned Prior
- Video Compression With Rate-Distortion Autoencoders
- Stochastic Adversarial Video Prediction
- Learning for Video Compression
- Disentangled Sequential Autoencoder
- Improving Inference for Neural Image Compression
- Feedback Recurrent AutoEncoder
- Variational Bayesian Quantization