AI Paper Atlas

Research map / Image generation / Image restoration

ZeST: Zero-Shot Material Transfer from a Single Image

Ta-Ying Cheng, Prafull Sharma, Andrew Markham, Niki Trigoni, Varun Jampani

ECCV 2024 · Image generation › Image restoration · approach: Image harmonization and color transformation · 30 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 depth-based ControlNet geometry guidance to IP-Adapter material transfer (ablation)ZeST · Qualitative (real images)geometry preservationImproved
initialize inpainting with foreground-grayscale image instead of original image (ablation)ZeST · Qualitative (real images)base color leakage and shadingImproved
initialize foreground with random noise instead of grayscale image (ablation)ZeST · Qualitative (real images)illumination correctnessHurt
use IP-Adapter material encoding instead of Dreambooth word token, with same guidance (ablation)ZeST · Synthetic datasetPSNR: 25.46 -> 25.82No clear effect
use IP-Adapter instead of Dreambooth on real images (ablation)ZeST · User study (real images)material fidelity: 3.25 -> 4.05Improved
add geometry and illumination guidance (replace Instruct-Pix2Pix pipeline) (ablation)IP-Adapter · Synthetic datasetPSNR: 16.92 -> 25.82Improved

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