Research map / Image generation / Fast diffusion sampling
Dark Noise Diffusion: Noise Synthesis for Low-Light Image Denoising
Liying Lu, Raphaël Achddou, Sabine E. Susstrunk
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 the two-branch architecture instead of UNet-only (ablation) | NoiseDiff (DDIM) · SID x100 | PSNR: 41.45 -> 42.03 | Improved |
| add the two-branch architecture instead of UNet-only (ablation) | NoiseDiff (DDIM) · SID x300 | PSNR: 36.59 -> 36.67 | Improved |
| add the two-branch architecture instead of UNet-only (ablation) | NoiseDiff (DDIM) · SID x100 noise KLD | KL divergence: 0.645 -> 0.622 | Improved |
| add positional encoding (ablation) | NoiseDiff (DDIM) · SID x100 | PSNR: 41.64 -> 42.03 | Improved |
| add positional encoding (ablation) | NoiseDiff (DDIM) · SID x300 | PSNR: 35.99 -> 36.67 | Improved |
| add positional encoding (ablation) | NoiseDiff (DDIM) · SID x100 noise KLD | KL divergence: 0.622 -> 0.622 | Improved |
| use DDPM instead of DDIM sampling (ablation) | NoiseDiff · SID x100 | PSNR: 42.03 -> 42.42 | Improved |
| use DDPM instead of DDIM sampling (ablation) | NoiseDiff · SID x300 | PSNR: 36.67 -> 36.87 | Improved |
| sample scenes regardless of camera setting for more diverse synthetic data (ablation) | NoiseDiff · SID x300 | PSNR: 36.87 -> 37.24 | Improved |
| sample scenes regardless of camera setting for more diverse synthetic data (ablation) | NoiseDiff · SID x250 | PSNR: 39.80 -> 39.67 | Mixed |
| add dark shading correction (ablation) | NoiseDiff · ELD x200 | PSNR: 42.25 -> 44.90 | Improved |
| add dark shading correction (ablation) | NoiseDiff · SID x250 | PSNR: 40.53 -> 41.26 | Improved |
Closest papers by meaning
- Noise Modeling in One Hour: Minimizing Preparation Efforts for Self-supervised Low-Light RAW Image Denoising (CVPR 2025 · similarity 0.90)
- Realistic Noise Synthesis with Diffusion Models (AAAI 2025 · similarity 0.89)
- One More Step: A Versatile Plug-and-Play Module for Rectifying Diffusion Schedule Flaws and Enhancing Low-Frequency Controls (CVPR 2024 · similarity 0.89)
- Warm Diffusion: Recipe for Blur-Noise Mixture Diffusion Models (ICLR 2025 · similarity 0.89)
- Dual-Rate Diffusion: Accelerating diffusion models with an interleaved heavy-light network (NeurIPS 2026 · similarity 0.88)
- Zero-Shot Low-Light Image Enhancement via Latent Diffusion Models (AAAI 2025 · similarity 0.88)
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.
