Research map / Image generation / Editing and style transfer
Diffusion-based Image Translation using disentangled style and content representation
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Use DiffuseIT instead of CLIP-GD | text-guided translation · Animals | SFID: 12.50 -> 9.98 | Helped |
| Use DiffuseIT instead of FlexIT | text-guided translation · Landscapes | CSFID: 60.04 -> 54.48 | Helped |
| Use DiffuseIT instead of FlexIT | text-guided translation · Animals | LPIPS: 0.215 -> 0.372 | Hurt |
| Use DiffuseIT instead of WCT2 | image-guided translation · Image-guided user study | Realism score: 4.64 -> 4.25 | Hurt |
| Add semantic divergence loss (ablation) | DiffuseIT · Animals | SFID: 15.00 -> 9.98 | Helped |
| Add semantic divergence loss (ablation) | DiffuseIT · Animals | CSFID: 53.43 -> 41.07 | Helped |
| Add content contrastive loss (ablation) | DiffuseIT · Animals | LPIPS: 0.477 -> 0.372 | Helped |
| Use the proposed content loss instead of LPIPS loss (ablation) | DiffuseIT · Animals | LPIPS: 0.400 -> 0.372 | Helped |
| Use semantic divergence loss instead of pixel-wise L2 maximization (ablation) | DiffuseIT · Animals | SFID: 13.18 -> 9.98 | Helped |
| Add the resampling strategy (ablation) | DiffuseIT · Animals | SFID: 11.88 -> 9.98 | Helped |
| Add the resampling strategy (ablation) | DiffuseIT · Animals | CSFID: 59.09 -> 41.07 | Helped |
| Use DINO ViT features instead of VGG16 features (ablation) | DiffuseIT · Animals | LPIPS: 0.518 -> 0.372 | Helped |
Closest papers by meaning
- Style Injection in Diffusion: A Training-free Approach for Adapting Large-scale Diffusion Models for Style Transfer (CVPR 2024 · similarity 0.91)
- Z*: Zero-shot Style Transfer via Attention Reweighting (CVPR 2024 · similarity 0.89)
- DEADiff: An Efficient Stylization Diffusion Model with Disentangled Representations (CVPR 2024 · similarity 0.89)
- SigStyle: Signature Style Transfer via Personalized Text-to-Image Models (AAAI 2025 · similarity 0.89)
- Noise Diffusion for Enhancing Semantic Faithfulness in Text-to-Image Synthesis (CVPR 2025 · similarity 0.88)
- Less is More: Masking Elements in Image Condition Features Avoids Content Leakages in Style Transfer Diffusion Models (ICLR 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.