AI Paper Atlas

Research map / Image generation / Text-to-image diffusion

Text-Aware Image Restoration with Diffusion Models

Jaewon Min, Jin Hyeon Kim, Paul Hyunbin Cho, Jaeeun Lee, Jihye Park, Park Min Kyu, Sangpil Kim, Hyunhee Park, Seungryong Kim

ICLR 2026 · Image generation › Text-to-image diffusion · approach: Compositional text-to-image guidance · 3 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 text prompts from the spotter at inference (null -> predicted) (ablation)TeReDiff stage 3 · SA-Text Level2end-to-end recognition (None): 23.46 -> 26.39Improved
use ground-truth text prompt instead of predicted (ablation)TeReDiff stage 3 · SA-Text Level2end-to-end recognition (None): 26.39 -> 33.31Improved
advance training from stage 1 to joint stage 3 (ablation)TeReDiff · SA-Text Level2end-to-end recognition (None): 21.24 -> 23.46 (null prompt)Improved
use diffusion features instead of ResNet-50 for text spotting (100K) (ablation)TESTR · SA-Text testend-to-end recognition (None): 17.03 -> 32.29Improved
use diffusion features instead of ResNet-50 with 20K data (ablation)TESTR · SA-Text testend-to-end recognition (None): 1.81 -> 27.42Improved
caption-style prompt instead of tag-style with predicted text (ablation)TeReDiff stage 3 · SA-Text Level2end-to-end recognition (None): 24.26 -> 26.39 (recall 59.25 -> 52.86)Mixed
caption vs tag prompt style with GT text (ablation)TeReDiff stage 3 · SA-Text Level2end-to-end recognition (None): 32.02 -> 33.31No clear effect
use TeReDiff instead of prior diffusion restorersTeReDiff · SA-Text Level3text detection F1: 50.65 -> 56.13 (vs DiffBIR)Improved
apply prior restorers to strongly degraded imagesSeeSR / DiffBIR etc. · SA-Text Level2/3end-to-end recognition: LQ 20.49 vs DiffBIR 21.64; Real-Text LQ 47.08 vs DiffBIR 39.27Hurt
use TeReDiff vs DiffBIR on image qualityTeReDiff · SA-Text testnon-reference metrics (MANIQA): LPIPS 0.3636 -> 0.2828; MANIQA 0.6771 -> 0.6471Mixed

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