AI Paper Atlas

Research map / Image generation / Editing and style transfer

ID-PreFeR: ID-Preserving Face Restoration with Mixed Data Quality

Chengxuan Zhu, Yuchen Hong, Qingnan Fan, Qi Zhang, Bingtao Fu, Jinxiu Liang, Jinwei Chen, Huaqi Zhang, Chao Xu, Boxin Shi

ECCV 2026 · Image generation › Editing and style transfer · approach: Diffusion face generation and editing · 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 latent prior preservation loss (ablation)ID-PreFeR · CelebRefMANIQA: 0.624 -> 0.649Improved
add latent prior preservation loss (ablation)ID-PreFeR · CelebRefID similarity: 0.746 -> 0.767Improved
add ID-quality disentanglement (ablation)ID-PreFeR · CelebRefMANIQA: 0.555 -> 0.649Improved
add ID-quality disentanglement (ablation)ID-PreFeR · CelebRefLPIPS: 0.279 -> 0.242Improved
add ID-quality disentanglement (ablation)ID-PreFeR · CelebRef, #HQ=1ID similarity: 0.7632 -> 0.7613No clear effect
add ID-preserving sampling (ablation)ID-PreFeR · CelebRefID similarity: 0.734 -> 0.767Improved
add ID-preserving sampling (ablation)ID-PreFeR · CelebRef, #HQ=5ID similarity: 0.7338 -> 0.7671Improved
add ID-preserving sampling (ablation)ID-PreFeR · Inference costinference time (s): 5.6 -> 29.0Hurt
add ID-preserving sampling (ablation)ID-PreFeR · Inference costVRAM (GB): 24 -> 34.8Hurt
restrict LoRA to decoder cross-attention instead of all cross-attention (ablation)ID-PreFeR · CelebRefparameters (M): 11.82 -> 6.41Improved
restrict LoRA to decoder cross-attention instead of all CA and SA (ablation)ID-PreFeR · CelebRefID similarity: 0.756 -> 0.767Improved
reduce number of HQ references from 5 to 1 (ablation)ID-PreFeR · CelebRefID similarity: 0.7671 -> 0.7613No clear effect

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