AI Paper Atlas

Research map / Image generation / Image restoration

When AWGN-Based Denoiser Meets Real Noises

Yuqian Zhou, Jianbo Jiao, Haibin Huang, Yang Wang, Jue Wang, Honghui Shi, Thomas Huang

AAAI 2020 · Image generation › Image restoration · approach: Raw and low-light noise modeling · 146 citations (OpenAlex)

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 changedWhereMeasuredEffect
Train with AWGN-RVIN and explicitly estimate the two noise typesProposed denoiser · Set20 AWGN-RVINPSNR: 32.09 -> 32.43; 29.97 -> 30.47; 29.52 -> 29.82; 27.90 -> 28.41Improved
Use the proposed blind model instead of DnCNN-BSignal-dependent noise denoising · CBSD68 (20, 10)PSNR: 34.38 -> 34.75Improved
Use the proposed blind model instead of DnCNN-BSignal-dependent noise denoising · CBSD68 (20, 20)PSNR: 31.72 -> 31.32Hurt
Use the proposed blind model instead of DnCNN-BSignal-dependent noise denoising · CBSD68 (40, 10) and (40, 20)PSNR: 32.08 -> 32.12; 30.32 -> 30.33Mixed
Apply PD adaptation to the AWGN-RVIN base modelProposed denoiser · DNDPSNR: 32.60 -> 38.40Improved
Apply PD adaptation to existing denoisersCBM3D, CDnCNN-B, and FFDNet · DNDPSNR: 34.51 -> 35.02; 32.43 -> 35.44; 34.40 -> 37.56Improved
Add RVIN to AWGN training (ablation)Proposed denoiser · Real noisy night photosdead-pixel removalImproved
Use stride s=1 rather than the full pipeline's s=2 (ablation)PD refinement · DNDPSNR: 38.40 -> 32.60Hurt
Use stride s=3 rather than the full pipeline's s=2 (ablation)PD refinement · DNDPSNR: 38.40 -> 37.90Hurt
Stop after pixel-shuffle upsampling (I), omitting later refinement (ablation)PD refinement · DNDPSNR: 38.40 -> 37.00Hurt
Use DI refinement instead of the full pipeline (ablation)PD refinement · DNDPSNR: 38.40 -> 37.20Hurt
Increase blending factor k (ablation)PD refinement · Real noisy imagesdetail preservationHurt

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy