Back to Search
Start Over
Improving Inference for Neural Image Compression
- Publication Year :
- 2020
-
Abstract
- We consider the problem of lossy image compression with deep latent variable models. State-of-the-art methods build on hierarchical variational autoencoders (VAEs) and learn inference networks to predict a compressible latent representation of each data point. Drawing on the variational inference perspective on compression, we identify three approximation gaps which limit performance in the conventional approach: an amortization gap, a discretization gap, and a marginalization gap. We propose remedies for each of these three limitations based on ideas related to iterative inference, stochastic annealing for discrete optimization, and bits-back coding, resulting in the first application of bits-back coding to lossy compression. In our experiments, which include extensive baseline comparisons and ablation studies, we achieve new state-of-the-art performance on lossy image compression using an established VAE architecture, by changing only the inference method.<br />9 pages + detailed supplement with additional results; various typos corrected. Camera-ready version paper at NeurIPS 2020
- Subjects :
- FOS: Computer and information sciences
Computer Science - Machine Learning
Statistics - Machine Learning
Image and Video Processing (eess.IV)
FOS: Electrical engineering, electronic engineering, information engineering
Machine Learning (stat.ML)
Data_CODINGANDINFORMATIONTHEORY
Electrical Engineering and Systems Science - Image and Video Processing
Machine Learning (cs.LG)
Subjects
Details
- Language :
- English
- Database :
- OpenAIRE
- Accession number :
- edsair.doi.dedup.....82e8d158ca9c81a2e10b979dbcb9377c