AI Paper Atlas

Research map / Image generation / Editing and style transfer

ArtFlow: Unbiased Image Style Transfer via Reversible Neural Flows

Jie An, Siyu Huang, Yibing Song, Dejing Dou, Wei Liu, Jiebo Luo

CVPR 2021 · Image generation › Editing and style transfer · approach: Arbitrary neural style transfer · 254 citations (OpenAlex)

Check 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 the VGG19 encoder-decoder with reversible PFN while using AdaIN transferAdaIN style transfer · Repeated style transfer (20 rounds)content leakImproved
Replace the VGG19 encoder-decoder with reversible PFN while using WCT transferWCT style transfer · Repeated style transfer (20 rounds)content leakImproved
Use PFN with AdaIN rather than vanilla AdaINAdaIN · Style transfer (512×512)Gram Loss: 0.00127 -> 0.00098Improved
Use PFN with WCT rather than vanilla WCTWCT · Style transfer (512×512)Gram Loss: 0.00074 -> 0.00078No clear effect
Use PFN with AdaIN rather than vanilla AdaINAdaIN · Style transfer (512×512)SSIM: 0.29 -> 0.45Improved
Use PFN with WCT rather than vanilla WCTWCT · Style transfer (512×512)SSIM: 0.27 -> 0.47Improved
Use PFN with AdaIN rather than vanilla AdaINAdaIN · Style transfer (512×512)Time (seconds): 0.064 -> 0.408Hurt
Use PFN with WCT rather than vanilla WCTWCT · Style transfer (512×512)Time (seconds): 0.997 -> 0.428Improved
Replace PFN (Flow8-Block2) with Flow4-Block4 (ablation)ArtFlow · Style transfer (512×512)SSIM: 0.45 -> 0.401Hurt
Replace PFN (Flow8-Block2) with Flow8-Block1 (ablation)ArtFlow · Style transfer (512×512)Gram Loss: 0.00098 -> 0.00139Hurt
Replace PFN (Flow8-Block2) with Flow16-Block1 (ablation)ArtFlow · Style transfer (512×512)Gram Loss: 0.00098 -> 0.00137Hurt
Replace PFN (Flow8-Block2) with Flow4-Block4 (ablation)ArtFlow · Style transfer (512×512)Gram Loss: 0.00098 -> 0.00130Hurt

Later papers that compared with this method

9 later accepted papers report ArtFlow as a baseline in their results tables; 8 of them report a better number on most of the results they share with it.

1 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