Research map / Image generation / Image restoration
Conditional Probability Models for Deep Image Compression
Fabian Mentzer, Eirikur Agustsson, Michael Tschannen, Radu Timofte, Luc Van Gool
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| Replace the uniform prior with the proposed 3D-CNN context model (ablation) | Fixed auto-encoder (L=6, K=16) · Kodak | rate: 0.646 bpp -> 0.579 bpp | Improved |
| Replace the uniform prior with a zeroth-order symbol histogram (ablation) | Fixed auto-encoder (L=6, K=16) · Kodak | rate: 0.646 bpp -> 0.642 bpp | Improved |
| Replace the uniform prior with a one-step conditional histogram (ablation) | Fixed auto-encoder (L=6, K=16) · Kodak | rate: 0.646 bpp -> 0.627 bpp | Improved |
| Use the proposed 3D-CNN context model instead of a first-order histogram (ablation) | Fixed auto-encoder (L=6, K=16) · Kodak | rate: 0.627 bpp -> 0.579 bpp | Improved |
| Train the context model jointly with the auto-encoder rather than separately after training (ablation) | Masked 3D-CNN context model and auto-encoder · Kodak | rate: 22% reduction in bpp | Improved |
| Clip the rate term using a target bitrate (ablation) | Proposed compression model · Image compression | performance | No clear effect |
Closest papers by meaning
- Joint Autoregressive and Hierarchical Priors for Learned Image Compression (NeurIPS 2018 · similarity 0.90)
- Entroformer: A Transformer-based Entropy Model for Learned Image Compression (ICLR 2022 · similarity 0.89)
- Robustly overfitting latents for flexible neural image compression (NeurIPS 2024 · similarity 0.89)
- Conditional Latent Coding with Learnable Synthesized Reference for Deep Image Compression (AAAI 2025 · similarity 0.88)
- Causal Context Adjustment Loss for Learned Image Compression (NeurIPS 2024 · similarity 0.88)
- Diversify, Contextualize, and Adapt: Efficient Entropy Modeling for Neural Image Codec (NeurIPS 2024 · similarity 0.87)
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.
