AI Paper Atlas

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

InstantRetouch: Personalized Image Retouching without Test-time Fine-tuning

Temesgen Muruts Weldengus, Binnan Liu, Fei Kou, Youwei Lyu, Jinwei Chen, Changqing Zou, Qingnan Fan

ECCV 2026 · Image generation › Text-to-image diffusion · approach: Multi-subject personalization

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 LoRA adaptation to SigLIPv2 style encoder instead of frozen (ablation)RefRetouch encoder · Ablation validation setPSNR (dB): 23.41 -> 32.40Improved
use SigLIPv2-LoRA instead of ResNet-50 encoder (ablation)RefRetouch encoder · Ablation validation setPSNR (dB): 30.00 -> 32.40Improved
use color-space MLP decoder instead of UNet decoder (ablation)RefRetouch decoder · Ablation validation setPSNR (dB): 29.58 -> 32.40Improved
use additive conditioning instead of AdaLN (ablation)RefRetouch decoder · Ablation validation setPSNR (dB): 31.93 -> 32.40Improved
use additive conditioning instead of cross-attention (ablation)RefRetouch decoder · Ablation validation setPSNR (dB): 22.66 -> 32.40Improved
retrieve top-3 relevant references instead of using all references (ablation)RAR · MIT-FiveK (20 references)PSNR (dB): 21.99 -> 22.95Improved
increase retrieved references from K=3 to K=5 (ablation)RAR · MIT-FiveK (20 references)PSNR (dB): 22.95 -> 23.37Improved
use LoRA-adapted SigLIPv2 instead of frozen SigLIPv2 for retrieval embeddings (ablation)RAR · MIT-FiveK (20 references)PSNR (dB): 21.02 -> 22.95Improved
increase number of references 1 to 4 on single-style benchmarkRefRetouch · VCIRBPSNR (dB): 29.13 -> 29.82No clear effect
increase references 1 to 6RefRetouch · PPR10K-groupsPSNR (dB): 22.51 -> 25.27Improved
retrain MSM on the paper's datasetMSM · MIT-FiveK reconstructionPSNR (dB): 28.94 -> 26.32Hurt
compare to per-user trained AdaInt/RSFNet on PPR10K usersRefRetouch (tuning-free) · PPR10K usersPSNR (dB): 21.75 vs 22.01 (20 refs); 22.47 vs 22.82 (100 refs)Mixed

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