Research map / Image generation / Editing and style transfer
CoCoDiff: Correspondence-Consistent Diffusion Model for Fine-grained Style Transfer
Wenbo Nie, Zixiang Li, Renshuai Tao, Bin WU, Yunchao Wei, Yao Zhao
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add Sobel content loss to Gram style loss (ablation) | CoCoDiff · WikiArt style transfer | FID: 23.471 -> 18.432 | Improved |
| add Gram style loss to Sobel content loss (ablation) | CoCoDiff · WikiArt style transfer | FID: 29.845 -> 18.432 | Improved |
| add Sobel loss alone (ablation) | CoCoDiff · WikiArt style transfer | CFSD: 0.804 -> 0.631 | Improved |
| add Sobel loss alone (ablation) | CoCoDiff · WikiArt style transfer | FID: 26.513 -> 29.845 | Hurt |
| add Gram loss alone (ablation) | CoCoDiff · WikiArt style transfer | LPIPS: 0.697 -> 0.753 | Hurt |
| add AdaIN tone harmonization (ablation) | CoCoDiff · WikiArt style transfer | FID: 20.515 -> 18.432 | Improved |
| add AdaIN tone harmonization (ablation) | CoCoDiff · WikiArt style transfer | LPIPS: 0.762 -> 0.549 | Improved |
| use injection weight w=0.6 instead of 0.3 (ablation) | CoCoDiff · WikiArt style transfer | FID: 23.717 -> 18.432 | Improved |
| use injection weight w=0.6 instead of 3 (ablation) | CoCoDiff · WikiArt style transfer | FID: 35.937 -> 18.432 | Improved |
| use adaptive early stopping instead of fixed 1-5 iterations (ablation) | CoCoDiff cycle module · 10 content and 10 style images | LPIPS and FID | Improved |
| swap backbone from SD v1.4 to SD v1.5 (ablation) | CoCoDiff · WikiArt style transfer | FID: 18.432 -> 17.995 | Improved |
| swap backbone from SD v1.4 to SDXL (ablation) | CoCoDiff · WikiArt style transfer | FID: 18.432 -> 21.412 | Hurt |
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.
- StyTR2 — StyTr2: Image Style Transfer With Transformers (CVPR 2022) · better on 4 of 4 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
- SigStyle: Signature Style Transfer via Personalized Text-to-Image Models (AAAI 2025 · similarity 0.90)
- Semantix: An Energy-guided Sampler for Semantic Style Transfer (ICLR 2025 · similarity 0.90)
- DEADiff: An Efficient Stylization Diffusion Model with Disentangled Representations (CVPR 2024 · similarity 0.89)
- OmniStyle: Filtering High Quality Style Transfer Data at Scale (CVPR 2025 · similarity 0.88)
- SCSA: A Plug-and-Play Semantic Continuous-Sparse Attention for Arbitrary Semantic Style Transfer (CVPR 2025 · similarity 0.88)
- OmniConsistency: Learning Style-Agnostic Consistency from Paired Stylization Data (NeurIPS 2025 · similarity 0.88)
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.
