AI Paper Atlas

Research map / Image generation / Image restoration

PGDiff: Guiding Diffusion Models for Versatile Face Restoration via Partial Guidance

NeurIPS 2023 · Image generation › Image restoration · approach: Diffusion-based universal restoration

Read the paperCheck your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's title only (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
Replace fixed degradation-model guidance with guidance on desired image propertiesPGDiff · Blind face restoration, inpainting, and colorizationFIDMixed
Add identity reference guidancePGDiff blind face restoration · CelebRef-HQ syntheticIDS: 0.44 -> 0.76; increasing by 0.32Helped
Add dynamic guidance weight instead of constant guidance scale (ablation)PGDiff face inpainting · Face inpaintingSimilarity and sharpness of unmasked regionsHelped
Use multiple gradient steps rather than one guidance step (ablation)PGDiff blind face restoration · Blind face restorationArtifact removal and detail quality: N=1, 2, and 3Helped
Add perceptual and adversarial loss guidancePGDiff blind face restoration · Blind face restorationFidelity and photo-realismHelped
Add identity loss guidanceCodeFormer with PGDiff guidance · Reference-based face restorationRecovery of personal detailsHelped

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