AI Paper Atlas

Research map / Image generation / Editing and style transfer

Warping the Residuals for Image Editing with StyleGAN

Ahmet Burak Yildirim, Hamza Pehlivan, Ayşegül Dündar

IJCV 2024 · Image generation › Editing and style transfer · approach: GAN inversion for editing · 8 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 trained feature-space flow with flow loss (vs StyleRes without warping)StyleRes · CelebA-HQ reconstructionFID: 7.04 -> 6.59Improved
add trained feature-space flow with flow lossStyleRes · CelebA-HQ pose editingId(+): 0.58 -> 0.74Improved
add trained feature-space flow with flow lossStyleRes · CelebA-HQ pose editingFID(+): 11.31 -> 10.40Improved
add flow estimation to StyleResStyleRes · CelebA-HQ runtimeruntime (sec): 0.0429 -> 0.1254Hurt
add flow guidance loss from pretrained flow network (ablation)trained flow WarpRes · CelebA-HQ pose editingFID(+): 16.38 -> 10.40Improved
add flow guidance loss from pretrained flow network (ablation)trained flow WarpRes · CelebA-HQ smile removalFID(-): 24.54 -> 21.52Improved
add flow guidance loss from pretrained flow network (ablation)trained flow WarpRes · CelebA-HQ reconstructionFID: 7.24 -> 6.59Improved
replace pretrained image-space flow with trained feature-space flow (ablation)WarpRes · CelebA-HQ reconstructionFID: 7.55 -> 6.59Improved
replace pretrained image-space flow with trained feature-space flow (ablation)WarpRes · CelebA-HQ pose editingId(+): 0.67 -> 0.74Improved
add WarpRes on top of pSp encoderpSp · CelebA-HQ reconstructionFID: 23.86 -> 6.60Improved
add WarpRes instead of StyleRes on pSp encoderpSp + StyleRes · CelebA-HQ pose editingId(+): 0.58 -> 0.72Improved
add WarpRes on top of StyleTransformerStyleTransformer · CelebA-HQ reconstructionFID: 21.82 -> 6.85Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 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