AI Paper Atlas

Research map / Image generation / Image restoration

CoordFill: Efficient High-Resolution Image Inpainting via Parameterized Coordinate Querying

Weihuang Liu, Xiaodong Cun, Chi-Man Pun, Menghan Xia, Yong Zhang, Jue Wang

AAAI 2023 · Image generation › Image restoration · approach: Image harmonization and color transformation · 64 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 CoordFill instead of LaMahigh-resolution inpainting · Places2 2048×2048speed: 598ms -> 26msImproved
Use CoordFill instead of ZITShigh-resolution inpainting · Unsplash 1024×1024PSNR: 32.532 -> 33.093Improved
Use CoordFill instead of LaMahigh-resolution inpainting · CelebA-HQ 1024×1024PSNR: 26.098 -> 28.756Improved
Use CoordFill instead of ZITSinpainting · Places2 1024×1024PSNR: 26.389 -> 26.322Hurt
Add masked-region-only supervision (ablation)CoordFill with Pixel-Query · Places2 512×512PSNR: 26.0434 -> 26.2160Improved
Use Pixel-Query instead of a convolutional decoder (ablation)CoordFill with masked prediction · Places2 512×512PSNR: 25.9557 -> 26.2160Improved
Use Pixel-Query instead of a shared MLP decoder (ablation)CoordFill with masked prediction · Places2 512×512PSNR: 24.6480 -> 26.2160Improved
Add masked-region-only supervision (ablation)CoordFill with D_Conv · Places2 512×512PSNR: 25.5420 -> 25.9557Improved
Add AttFFC instead of using ResFFC (ablation)CoordFill with Pixel-Query and masked prediction · Places2 512×512PSNR: 26.2160 -> 26.3177Improved
Add resolution conditioning (ablation)CoordFill with Pixel-Query and AttFFC · Places2 512×512PSNR: 26.3177 -> 26.3653Improved

Later papers that compared with this method

1 later accepted paper reports Coordfill as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 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