AI Paper Atlas

Research map / Image generation / Image restoration

Learning to Corrupt for Better Restoration

Joonkyu Park, Wooseok Lee, Jaeha Kim, Sehoon Kim, Bokyeung Lee, Kyoung Mu Lee

ECCV 2026 · Image generation › Image restoration · approach: Diffusion-based universal restoration · 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
predict input-aware timestep instead of fixing it (ablation)IAC-IR · RealSRLIQE: 4.571 (t=400) -> 4.668Improved
larger fixed timestep (ablation)IAC-IR · RealSRPSNR vs LIQE: PSNR 23.064 -> 20.492; LIQE 3.816 -> 4.571Mixed
use ground-truth timestep tGT instead of predicted (ablation)IAC-IR · RealSRPSNR: 22.509 -> 22.698Improved
add explicit noise supervision L_eps (ablation)IAC-IR · RealSRMANIQA: 0.483 -> 0.589Improved
replace L_eps with KL divergence noise loss (ablation)IAC-IR · RealSRMANIQA: 0.589 -> 0.523Hurt
add timestep loss Lt (ablation)IAC-IR · RealSRMANIQA: 0.457 -> 0.483Improved
use both Lt and L_eps together (ablation)IAC-IR · RealSRMANIQA: 0.457 -> 0.589Improved
distillation with predicted timestep and noise instead of random (ablation)Ldistill · RealSRMANIQA: 0.479 -> 0.589Improved
use Gaussianity measure G() instead of KL (ablation)IAC-IR · RealSRPSNR: 21.190 -> 22.509Improved
increase iterations from 1 to 5 (ablation)IAC-IR · RealSRPSNR / MANIQA: PSNR 22.509 -> 18.798; MANIQA 0.589 -> 0.635Mixed
IAC-IR vs prior one-step diffusion IR on perceptual metricsIAC-IR · RealSRLIQE: 4.068 (OSEDiff) -> 4.668Improved
IAC-IR vs OSEDiff on distortionIAC-IR · RealSRPSNR: 23.589 -> 22.509Hurt

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