AI Paper Atlas

Research map / Image generation / Image restoration

YeTI: You Only Need Two Noisy Images for Real-World sRGB Noise Generation

Jaekyun Ko, Byung Wan Lim, Dongjin Kim, Soomin Lee, Tae Hyun Kim

ECCV 2026 · Image generation › Image restoration · approach: Raw and low-light noise modeling · 0 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 disentangled structure+noise conditioning in C-DiT (vs single encoder) (ablation)YeTI C-DiT · MAI2021KLD: 0.1584 -> 0.0793Improved
add L_Norm on noise latent (ablation)YeTI RAE · SIDD val metadata classificationTop-1 accuracy (%): 47.20 -> 65.22Improved
add contrastive projection head (ablation)YeTI RAE · SIDD val metadata classificationTop-1 accuracy (%): 51.20 -> 65.22Improved
add contrastive loss L_Cont (ablation)YeTI RAE · SIDD val structure purityPSNR: 15.49 -> 23.06Improved
add projection head (ablation)YeTI RAE · SIDD val structure purityPSNR: 17.21 -> 23.06Improved
add total variation loss L_TV (ablation)YeTI RAE · SIDD val structure purityPSNR: 21.49 -> 23.06Improved
use subtractive-additive injection instead of element-wise addition (ablation)YeTI RAE · SIDD val structure purityPSNR: 12.70 -> 23.06Improved
use subtractive-additive injection instead of concatenation (ablation)YeTI RAE · SIDD val structure purityPSNR: 14.42 -> 23.06Improved
mix 50% YeTI synthetic with real noisy images for trainingAP-BSN · SIDD val denoisingPSNR: 36.86 -> 37.09Improved
mix 50% YeTI synthetic with real noisy images for trainingAP-BSN · DND denoisingPSNR: 38.37 -> 38.55Improved
raise lambda_TV and lambda_Norm from 1e-6 to 1e-5 (ablation)YeTI RAE · SIDD val metadata classificationTop-1 accuracy (%): 65.22 -> 42.71Hurt
shift burst frames by N=4 pixels during training (ablation)YeTI · SIDD valKLD: 0.0256 -> 0.0416Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

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

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