Research map / Probabilistic and scientific modeling / Variational and generative sampling
Improving Inference for Neural Image Compression
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add stochastic Gumbel annealing and iterative inference at compression time | Base Hyperprior · Kodak | BD rate savings vs Base Hyperprior (%): 15% | Helped |
| add lossy bits-back coding on top of SGA (ablation) | SGA · Kodak | BD rate savings vs Base Hyperprior (%): 15% -> 17% | Helped |
| use SGA instead of straight-through estimator, MAP, uniform noise or deterministic annealing for discretization (ablation) | SGA · Kodak | true rate-distortion loss / BD rate | Helped |
| add iterative inference and SGA before bits-back coding (ablation) | lossy bits-back · Kodak | BD rate savings | Helped |
Closest papers by meaning
- Robustly overfitting latents for flexible neural image compression (NeurIPS 2024 · similarity 0.92)
- Hierarchical Quantized Autoencoders (NeurIPS 2020 · similarity 0.91)
- Compressing Images by Encoding Their Latent Representations with Relative Entropy Coding (NeurIPS 2020 · similarity 0.91)
- Variational image compression with a scale hyperprior (ICLR 2018 · similarity 0.91)
- Practical lossless compression with latent variables using bits back coding (ICLR 2019 · similarity 0.91)
- Bound-Conditioned Latent Inference for Progressive Image Compression (NeurIPS 2026 · similarity 0.90)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.