AI Paper Atlas

Research map / Video / Video generation

Refacade: Editing Object with Given Reference Texture

Youze Huang, Penghui Ruan, Bojia Zi, Xianbiao Qi, Jianan Wang, Rong Xiao

CVPR 2026 · Video › Video generation · approach: Diffusion-based video editing

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
replace Canny/HED/gray/depth structure conditions with texture-remover untextured video (ablation)Refacade stage 1 · Video benchmarkforeground DINO score: 0.1906 (Canny) -> 0.2622Improved
add jigsaw permutation of the reference image (ablation)Refacade stage 1 (Canny) · Video benchmarkbackground MSE: 68.83 -> 30.65Improved
add jigsaw permutation of the reference image (ablation)Refacade stage 1 (Canny) · Video benchmarkforeground LPIPS: 0.7674 -> 0.7347Improved
add jigsaw permutation of the reference image (ablation)Refacade stage 1 (Canny) · Video benchmarkGPT-5 score: 2.10 -> 2.42Improved
add stage-2 high-quality finetuning on Pexels (ablation)Refacade · Video benchmarkforeground DINO score: 0.2622 -> 0.3241Improved
add stage-2 high-quality finetuning on Pexels (ablation)Refacade · Video benchmarkEWarp: 1.3510 -> 1.4248Hurt
use small jigsaw patches (2%) instead of 5-10% (ablation)Refacade jigsaw · Video benchmarkforeground DINO score: 0.2615 (10%) -> 0.2158 (2%)Hurt
use large jigsaw patches (50-100%) instead of 5-10% (ablation)Refacade jigsaw · Video benchmarkEWarp / texture fidelity: EWarp 1.4495 (10%) -> 1.3970 (100%); DINO 0.2615 -> 0.2380Mixed
jigsaw patch size changed between 2% and 100% (ablation)Refacade jigsaw · Video benchmarkbackground LPIPS: 0.0326 for all patch sizesNo clear effect
distill texture remover with DMD2 from 50 to 3 sampling stepsTexture remover · Inference costsampling steps: 50 -> 3 stepsImproved
use Refacade instead of general-purpose and inpainting editorsObject Retexture · Image benchmarkuser preference: 0.8944 vs 0.2236 (Flux-Kontext-Image)Improved

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