Research map / Image generation / Editing and style transfer
CCPL: Contrastive Coherence Preserving Loss for Versatile Style Transfer
Zijie Wu, Zhen Zhu, Junping Du, Xiang Bai
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 CCPL to SCTNet (ablation) | SCTNet · Video/artistic style transfer (MS-COCO+WikiArt trained, MPI Sintel) | temporal loss (i=10): 12.22 -> 5.08 | Improved |
| add CCPL to SCTNet (ablation) | SCTNet · Video/artistic style transfer (MS-COCO+WikiArt trained, MPI Sintel) | LPIPS (i=10): 0.446 -> 0.367 | Improved |
| add CCPL to SCTNet (ablation) | SCTNet · Video/artistic style transfer (MS-COCO+WikiArt trained, MPI Sintel) | SIFID: 2.29 -> 2.43 | Hurt |
| add CCPL to SCTNet (ablation) | SCTNet · Video/artistic style transfer (MS-COCO+WikiArt trained, MPI Sintel) | human preference (Vid): 0.060 -> 0.138 | Improved |
| add CCPL to AdaIN (ablation) | AdaIN · Video/artistic style transfer (MS-COCO+WikiArt trained, MPI Sintel) | temporal loss (i=10): 7.92 -> 6.72 | Improved |
| add CCPL to SANet (ablation) | SANet · Video/artistic style transfer (MS-COCO+WikiArt trained, MPI Sintel) | temporal loss (i=10): 13.72 -> 7.09 | Improved |
| add CCPL to Linear (ablation) | Linear · Video/artistic style transfer (MS-COCO+WikiArt trained, MPI Sintel) | temporal loss (i=10): 7.61 -> 6.96 | Improved |
| add CCPL to AdaIN, SANet and Linear (ablation) | AdaIN/SANet/Linear · Video/artistic style transfer (MS-COCO+WikiArt trained, MPI Sintel) | SIFID: 2.44 -> 2.58; 2.40 -> 2.60; 2.38 -> 2.47 | Hurt |
| add L_nor instead of CCPL to SCTNet (ablation) | SCTNet · Video/artistic style transfer (MS-COCO+WikiArt trained, MPI Sintel) | temporal loss (i=10): 12.22 -> 11.54 | No clear effect |
| add CCPL to SCTNet (ablation) | SCTNet · Photo-realistic style transfer | LPIPS: 0.427 -> 0.351 | Improved |
| add CCPL to SCTNet (ablation) | SCTNet · Photo-realistic style transfer | SIFID: 1.65 -> 2.14 | Hurt |
| add CCPL to SCTNet (ablation) | SCTNet · Photo-realistic style transfer | human preference: 0.128 -> 0.262 | Improved |
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.
- AdaAttN — AdaAttN: Revisit Attention Mechanism in Arbitrary Neural Style Transfer (ICCV 2021) · better on 1 of 1 shared result
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
- Coherent Online Video Style Transfer (ICCV 2017 · similarity 0.91)
- Arbitrary Video Style Transfer via Multi-Channel Correlation (AAAI 2021 · similarity 0.89)
- Learning Linear Transformations for Fast Image and Video Style Transfer (CVPR 2019 · similarity 0.89)
- A Content Transformation Block for Image Style Transfer (CVPR 2019 · similarity 0.88)
- CLeaR: A Unified Framework for Resolving the Leakage–Degradation Dilemma in Style Transfer (NeurIPS 2026 · similarity 0.87)
- OmniStyle: Filtering High Quality Style Transfer Data at Scale (CVPR 2025 · similarity 0.86)
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.
