Research map / Image generation / Editing and style transfer
CLIPstyler: Image Style Transfer With a Single Text Condition
Check 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 |
|---|---|---|---|
| add PatchCLIP loss (ablation) | CLIPstyler · Text-guided manipulation ablation user study | Preference Score: 1.69 vs 3.92 | Helped |
| add threshold rejection (ablation) | CLIPstyler · Text-guided manipulation ablation user study | Preference Score: 3.02 vs 3.92 | Helped |
| add perspective augmentation (ablation) | CLIPstyler · Text-guided manipulation ablation user study | Preference Score: 2.61 vs 3.92 | Helped |
| add directional CLIP loss (ablation) | CLIPstyler · Text-guided manipulation ablation user study | Preference Score: 2.58 vs 3.92 | Helped |
| use perspective augmentation instead of random affine transformation (ablation) | CLIPstyler · Text-guided style transfer | perceptual quality | Helped |
| crop patches from content images as well as output images (ablation) | CLIPstyler · Text-guided style transfer | perceptual quality | No clear effect |
| use different patch crop sizes (ablation) | CLIPstyler · Text-guided style transfer | style pattern scale | Mixed |
| use CLIPstyler instead of the strongest baseline, AdaIN+CLIP | text-guided manipulation models · Text-guided manipulation user study | Style: 3.32 vs 3.78 | Helped |
| use CLIPstyler instead of the strongest baseline, StyleCLIP | text-guided manipulation models · Text-guided manipulation user study | Content: 4.35 vs 4.18 | Hurt |
| use CLIPstyler instead of AdaIN+CLIP | text-guided manipulation models · Text-guided manipulation patch-wise evaluation | CLIP score: 0.2487 vs 0.2515 | Helped |
Closest papers by meaning
- SigStyle: Signature Style Transfer via Personalized Text-to-Image Models (AAAI 2025 · similarity 0.88)
- StyleStudio: Text-Driven Style Transfer with Selective Control of Style Elements (CVPR 2025 · similarity 0.88)
- ArtAdapter: Text-to-Image Style Transfer using Multi-Level Style Encoder and Explicit Adaptation (CVPR 2024 · similarity 0.87)
- Style Injection in Diffusion: A Training-free Approach for Adapting Large-scale Diffusion Models for Style Transfer (CVPR 2024 · similarity 0.87)
- CSGO: Content-Style Composition in Text-to-Image Generation (NeurIPS 2025 · similarity 0.87)
- SCSA: A Plug-and-Play Semantic Continuous-Sparse Attention for Arbitrary Semantic Style Transfer (CVPR 2025 · similarity 0.87)
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.