Research map / Image generation / Image restoration
StableCodec: Taming One-Step Diffusion for Extreme Image Compression
Tianyu Zhang, Xin Luo, Li Li, Dong Liu
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 |
|---|---|---|---|
| add auxiliary encoder (frozen ELIC analysis transform) (ablation) | StableCodec · Kodak | BD-rate DISTS: 0 -> -28.13% | Improved |
| add auxiliary decoder on top of auxiliary encoder (ablation) | StableCodec · Kodak | BD-rate DISTS: -28.13% -> -54.89% | Improved |
| increase LoRA ranks to 16/32 (ablation) | StableCodec · Kodak | BD-rate DISTS: 0 -> -17.96% | Improved |
| increase LoRA ranks beyond 16/32 to 32/64 (ablation) | StableCodec · Kodak | BD-rate DISTS: -17.96% -> -17.34% | Hurt |
| add LoRA to the VAE decoder (ablation) | StableCodec · Kodak | BD-rate PSNR vs DISTS: PSNR -7.12% -> -21.77%; DISTS -17.96% -> -5.13% | Mixed |
| add two-stage implicit bitrate pruning (IBP) (ablation) | StableCodec · Kodak | BD-rate PSNR: 0 -> -24.12% | Improved |
| add GAN loss in the second stage (ablation) | StableCodec · Kodak | BD-rate PSNR vs DISTS: PSNR -24.12% -> 24.41%; DISTS -16.60% -> -49.24% | Mixed |
| add CLIP distance loss (ablation) | StableCodec · Kodak | BD-rate PSNR: 24.41% -> 13.29% | Improved |
| use one-step diffusion with deep compression latent codec instead of multi-step diffusion | StableCodec vs DiffEIC · Kodak | decoding time (s): 7.423 -> 0.326 | Improved |
Later papers that compared with this method
5 later accepted papers report StableCodec as a baseline in their results tables; 4 of them report a better number on most of the results they share with it.
- DiT-IC: Aligned Diffusion Transformer for Efficient Image Compression (CVPR 2026 · better on 7 of 7 shared results)
- Ultra-Low Bitrate Perceptual Image Compression with Shallow Encoder (CVPR 2026 · better on 3 of 3 shared results)
- Training-Free Rate-Distortion-Perception Traversal With Diffusion (ICML 2026 · better on 0 of 1 shared result)
- CoD-Lite: Real-Time Diffusion-Based Generative Image Compression (ICML 2026 · better on 1 of 1 shared result)
- Autoregressive-based Progressive Coding for Ultra-Low Bitrate Image Compression (ICLR 2026 · better on 1 of 1 shared result)
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- OSCAR: One-Step Diffusion Codec Across Multiple Bit-rates (NeurIPS 2025 · similarity 0.91)
- Lossy Compression with Pretrained Diffusion Models (ICLR 2025 · similarity 0.91)
- Ultra Lowrate Image Compression with Semantic Residual Coding and Compression-aware Diffusion (ICML 2025 · similarity 0.90)
- Decouple Distortion from Perception: Region Adaptive Diffusion for Extreme-low Bitrate Perception Image Compression (CVPR 2025 · similarity 0.90)
- Progressive Compression with Universally Quantized Diffusion Models (ICLR 2025 · similarity 0.90)
- Bridging the Gap between Gaussian Diffusion Models and Universal Quantization for Image Compression (CVPR 2025 · 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.
