AI Paper Atlas

Research map / Image generation / Image restoration

Estimating Fine-Grained Noise Model via Contrastive Learning

CVPR 2022 · Image generation › Image restoration · approach: Self-supervised image denoising · 26 citations (OpenAlex)

Check 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 changedWhereMeasuredEffect
Use the proposed testing-image-only noise synthesis pipeline instead of CANGANSIDD noise synthesis · SIDD, average across camerasKL divergence: 0.0220 -> 0.0211Helped
Use the proposed noise synthesis pipeline instead of paired training dataUNet denoising · SIDD, S6, AllPSNR: 50.10 -> 50.13Helped
Use the proposed noise synthesis pipeline instead of paired training dataUNet denoising · SIDD, S6, AllSSIM: 0.9902 -> 0.9891Hurt
Add contrastive loss to the noise estimation model (ablation)Proposed pipeline · SIDD, S6, AllPSNR: 49.03 -> 50.13Helped
Add contrastive loss to the noise estimation model (ablation)Proposed pipeline · SIDD, S6, AllSSIM: 0.9868 -> 0.9891Helped
Use the fine-grained noise model instead of Hetero-G (ablation)Proposed pipeline · SIDD, S6, AllPSNR: 50.04 -> 50.13Helped
Use the fine-grained noise model instead of Hetero-G (ablation)Proposed pipeline · SIDD, S6, AllSSIM: 0.9874 -> 0.9891Helped

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