AI Paper Atlas

Research map / Image generation / Editing and style transfer

Eta Inversion: Designing an Optimal Eta Function for Diffusion-based Real Image Editing

Wonjun Kang, Kevin Galim, Hyung Il Koo

ECCV 2024 · Image generation › Editing and style transfer · approach: Inversion-based diffusion editing · 10 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 Eta Inversion (injecting real noise via a designed eta in DDIM sampling) instead of Direct InversionPtP on PIE-Bench · PIE-Bench, PtPCLIP similarity: 0.3091879120469093 -> 0.31247970819473Improved
use Eta Inversion (2) instead of Direct InversionPnP on PIE-Bench · PIE-Bench, PnPCLIP similarity: 0.3131952231909548 -> 0.3162599997861045Improved
use region-dependent (masked) eta instead of global eta (ablation)Eta Inversion (1) · PIE-Bench, PtPBG-LPIPS: 0.06737695838071626 -> 0.06568354647575429Improved
use region-dependent (masked) eta instead of global eta (ablation)Eta Inversion (2) · PIE-Bench, PtPCLIP similarity: 0.3126584818959236 -> 0.31247970819473 CLIP; DINO 0.018496008312795312 -> 0.017007591246760316Mixed
increase eta (EtaInv (2) vs (1)) to promote more editing (ablation)Eta Inversion · PIE-Bench, PtPCLIP similarity vs structural distance: CLIP 0.31007335275411 -> 0.31247970819473; DINO 0.013433658024296165 -> 0.017007591246760316Mixed
use larger eta without masking for style transferEta Inversion (3) · PIE-Bench change-style, PtPCLIP similarity: 0.31713469438254 -> 0.3285034172236919Improved
use larger eta for style transferEta Inversion (3) · PIE-Bench change-style, PtPDINO structural distance: 0.016382710146717726 -> 0.04185123109491542Hurt
use a decreasing linear eta schedule (ablation)Eta Inversion · PIE-Benchediting qualityImproved

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