AI Paper Atlas

Research map / Image generation / Image restoration

Learning Physics-Informed Noise Models from Dark Frames for Low-Light Raw Image Denoising

Hansen Feng, Lizhi Wang, Yiqi Huang, Yuzhi Wang, Lin Zhu, Hua Huang

TPAMI 2026 · Image generation › Image restoration · approach: Raw and low-light noise modeling · 5 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
add physics-guided noise decoupling (ablation)PNNP · ELD SonyPSNR: 44.26 -> 46.39Improved
add physics-guided noise decoupling (ablation)PNNP · SID SonyPSNR: 38.45 -> 40.83Improved
use 1x1 instead of 3x3 convolutions in PPM (ablation)PNNP · ELD SonyPSNR: 44.69 -> 46.39Improved
use 1x1 instead of 3x3 convolutions in PPM (ablation)PNNP · SID SonyPSNR: 39.68 -> 40.83Improved
use ISO-aware dual-branch structure instead of single branch (ablation)PNNP · ELD SonyPSNR: 45.89 -> 46.39Improved
use physics-aware PPM instead of flow-based modules (ablation)PNNP · ELD SonyPSNR: 46.00 -> 46.39Improved
add quantile loss to DDL (ablation)PNNP · ELD SonyPSNR: 46.05 -> 46.39Improved
add CDF loss to DDL (ablation)PNNP · ELD SonyPSNR: 46.16 -> 46.39Improved
add quantile loss to DDL (ablation)PNNP · Pixel-wise noiseKLD: 0.00203 -> 0.00048Improved
learn noise model from dark frames (PNNP) instead of paired real data (Paired Data)denoiser · ELD SonyPSNR: 43.22 -> 46.39Improved
PNNP versus paired real data on high-quality LRIDdenoiser · LRID-IndoorPSNR: 44.60 -> 44.69No clear effect
PNNP versus paired real data on LRID-Outdoordenoiser · LRID-OutdoorPSNR: 44.33 -> 44.17No clear effect

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