Research map / Image generation / Editing and style transfer
Content and Style Disentanglement for Artistic Style Transfer
Dmytro Kotovenko, Artsiom Sanakoyeu, Sabine Lang, Bjorn Ommer
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 fixpoint triplet style loss (ablation) | Ours · Style transfer ablation (Places365 photos) | deception rate: 0.45 -> 0.562 | Improved |
| add fixpoint triplet style loss (ablation) | Ours · Style transfer ablation (Places365 photos) | true style divergence (lower better): 1.14 -> 0.21 | Improved |
| add fixpoint triplet style loss (ablation) | Ours · Style transfer ablation (Places365 photos) | classification accuracy: 0.16 -> 0.17 | No clear effect |
| add fixpoint disentanglement loss (ablation) | Ours · Style transfer ablation (Places365 photos) | classification accuracy (content preservation): 0.08 -> 0.17 | Improved |
| add fixpoint disentanglement loss (ablation) | Ours · Style transfer ablation (Places365 photos) | deception rate: 0.52 -> 0.562 | Improved |
| add fixpoint disentanglement loss (ablation) | Ours · Style transfer ablation (Places365 photos) | true style divergence (lower better): 0.32 -> 0.21 | Improved |
| replace AST with proposed model (both learn style from image collection) | AST · Artist style patches (10 styles) | deception rate: 0.45 -> 0.562 | Improved |
| replace AST with proposed model | AST · Artist style patches (10 styles) | expert preference rate: 0.329 -> 0.486 | Improved |
| replace AST with proposed model | AST · Artist style patches (10 styles) | non-expert deception rate: 0.050 -> 0.182 | Improved |
| replace AST with proposed model | AST · Artist style patches (10 styles) | expert deception rate: 0.122 -> 0.240 | Improved |
| replace AST with proposed model | AST · Style transfer ablation (Places365 photos) | classification accuracy: 0.09 -> 0.17 | Mixed |
Closest papers by meaning
- Through Van Gogh’s Eyes: Global Style Transfer with Diffusion Model (ECCV 2026 · similarity 0.89)
- A Style-Aware Content Loss for Real-time HD Style Transfer (ECCV 2018 · similarity 0.89)
- DualAST: Dual Style-Learning Networks for Artistic Style Transfer (CVPR 2021 · similarity 0.88)
- ArtBank: Artistic Style Transfer with Pre-trained Diffusion Model and Implicit Style Prompt Bank (AAAI 2024 · similarity 0.88)
- Z*: Zero-shot Style Transfer via Attention Reweighting (CVPR 2024 · similarity 0.86)
- Conditional Balance: Improving Multi-Conditioning Trade-Offs in Image Generation (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.
