AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

BrushNet: A Plug-and-Play Image Inpainting Model with Decomposed Dual-Branch Diffusion

Xuan Ju, Xian Liu, Xintao Wang, Yuxuan Bian, Ying Shan, Qiang Xu

ECCV 2024 · Image generation › Fast diffusion sampling · approach: Diffusion models as perception tools · 172 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 dual-branch design instead of single-branch SD Inpainting (ablation)SD Inpainting · BrushBenchImageReward (x10): 11.00 -> 11.59Improved
fine-tune the base UNet in the dual-branch design (ablation)BrushNet · BrushBenchPSNR: 19.86 -> 20.13Improved
use VAE encoder instead of random conv layers for masked image (ablation)BrushNet · BrushBenchPSNR: 14.89 -> 19.86Improved
include mask in branch input (ablation)BrushNet · BrushBenchPSNR: 17.96 -> 19.86Improved
remove text cross-attention from the extra branch (ablation)BrushNet · BrushBenchLPIPS (x10^2, lower better): 19.44 -> 16.09Improved
add full UNet feature layer-by-layer instead of half (ablation)BrushNet · BrushBenchPSNR: 19.01 -> 19.86Improved
use full UNet feature addition instead of ControlNet-style addition (ablation)BrushNet · BrushBenchLPIPS (x10^2, lower better): 24.09 -> 16.09Improved
add blurred-mask blending in pixel space (ablation)BrushNet · BrushBenchPSNR: 19.86 -> 29.88Improved
blurred blending instead of direct paste (ablation)BrushNet · BrushBenchCLIP Sim: 26.80 -> 26.81Mixed
add blending operation to BrushNet at inferenceBrushNet · BrushBench outside-inpaintingCLIP Sim: 27.33 -> 27.22Hurt
replace ControlNet-Inpainting with BrushNetControlNet-Inpainting · BrushBench inside-inpaintingPSNR: 12.39 -> 21.65Improved

Later papers that compared with this method

16 later accepted papers report BrushNet as a baseline in their results tables; 15 of them report a better number on most of the results they share with it.

8 more not listed.

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