AI Paper Atlas

Research map / Image generation / Editing and style transfer

Industrial Style Transfer With Large-Scale Geometric Warping and Content Preservation

Jinchao Yang, Fei Guo, Shuo Chen, Jun Li, Jian Yang

CVPR 2022 · Image generation › Editing and style transfer · approach: Arbitrary neural style transfer · 16 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
Use LGW instead of GTST for large-scale geometric warpingGeometric warping · Geometric warpingmIoU: 0.7285 -> 0.9284Improved
Use LGW instead of DST for large-scale geometric warpingGeometric warping · Geometric warpingmIoU: 0.6000 -> 0.9284Improved
Add interest regularization (IR) to AdaIN (ablation)AdaIN · Stylization content preservationSSIM: 0.3424 -> 0.3886Improved
Add interest regularization (IR) to LinearWCT (ablation)LinearWCT · Stylization content preservationSSIM: 0.4612 -> 0.4932Improved
Add interest regularization (IR) to ArtFlow (ablation)ArtFlow · Stylization content preservationSSIM: 0.5042 -> 0.5643Improved
Add interest regularization (IR) to AdaIN (ablation)AdaIN · Stylization content preservationTime(s): 0.054 -> 0.054No clear effect
Add interest regularization (IR) to LinearWCT (ablation)LinearWCT · Stylization content preservationTime(s): 0.419 -> 0.416Improved
Add interest regularization (IR) to ArtFlow (ablation)ArtFlow · Stylization content preservationTime(s): 0.138 -> 0.140Hurt
Add position embedding to mask RAFT features (ablation)LGW · Geometric warpingWarping qualityImproved
Use mask inputs instead of RGB inputs for RAFT warpingLGW · Geometric warpingWarping loss and convergenceImproved

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