AI Paper Atlas

Research map / Image generation / Image restoration

FoundIR: Unleashing Million-scale Training Data to Advance Foundation Models for Image Restoration

Hao Li, Xiang Chen, Jiangxin Dong, Jinhui Tang, Jinshan Pan

ICCV 2025 · Image generation › Image restoration · approach: Efficient restoration architectures · 28 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
use incremental learning from isolated to coupled degradations instead of Mix-Train (ablation)FoundIR generalist · FoundIR benchmarkPSNR: 22.57 -> 23.70Improved
use incremental learning instead of Combine-Train (ablation)FoundIR generalist · FoundIR benchmarkPSNR: 23.40 -> 23.70Improved
use incremental learning instead of Sequence-Train (ablation)FoundIR generalist · FoundIR benchmarkPSNR: 22.71 -> 23.70Improved
order incremental learning isolated first then coupled rather than coupled then isolated (ablation)FoundIR generalist · FoundIR benchmarkPSNR: 22.33 -> 23.70Improved
add specialist models after the generalist (G -> S) versus generalist only (ablation)FoundIR · FoundIR benchmark L+HPSNR: 18.19 -> 22.82Improved
add generalist before specialist versus specialist only (ablation)FoundIR · FoundIR benchmark L+HPSNR: 13.06 -> 22.82Improved
apply specialist first then generalist (S -> G) instead of G -> S (ablation)FoundIR · FoundIR benchmark L+HPSNR: 22.82 -> 15.63Hurt
chain two specialists (S1 -> S2) instead of generalist then specialist (ablation)FoundIR · FoundIR benchmark L+HPSNR: 22.82 -> 13.52Hurt
train on 10% of the proposed dataset instead of 120,000 public real-world pairs (ablation)FoundIR · FoundIR benchmarkPSNR: 2.59dB improvement in PSNRImproved
increase training data from 10% to 100% of the proposed dataset (ablation)FoundIR · FoundIR benchmarkPSNRImproved
set degradation-agnostic coefficient gamma_T to 0.3 versus other values or none (ablation)FoundIR generalist · FoundIR benchmark (20 degradation types)PSNRImproved
train DiffUIR on the proposed dataset instead of public datasetsDiffUIR · RealRain-1KPSNR: 6dB PSNR gainImproved

Later papers that compared with this method

3 later accepted papers report FoundIR as a baseline in their results tables; 3 of them report a better number on most of the results they share with it.

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 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