Research map / Image generation / Text-to-image diffusion
FontStudio: Shape-Adaptive Diffusion Model for Coherent and Consistent Font Effect Generation
Xinzhi Mu, Li Chen, Bohan CHEN, Shuyang Gu, Jianmin Bao, Dong Chen, Ji Li, Yuhui Yuan
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 shape-adaptive attention to the shape-adaptive generation model (ablation) | SGM · GenerativeFont benchmark | M-CLIP-Ext: 22.07 -> 18.11 | Improved |
| add shape-adaptive attention (foreground content) (ablation) | SGM · GenerativeFont benchmark | M-CLIP-Int: 27.10 -> 27.26 | Improved |
| train with a segmentation-mask condition instead of depth-map conditions (ablation) | SGM · GenerativeFont benchmark | M-CLIP-Int: 24.51 -> 27.26 | Improved |
| use ControlNet-Canny or Depth trained on rectangle canvases instead of the shape-adaptive model | SDXL-ControlNet · GenerativeFont benchmark | M-CLIP-Ext: 21.52 and 23.24 vs 18.11 | Hurt |
| add shape-adaptive effect transfer (SAET) across letters (ablation) | FontStudio · GenerativeFont benchmark | DINO: 54.27 -> 67.07 | Improved |
| replace StyleAligned with SAET (ablation) | FontStudio · GenerativeFont benchmark | DINO: 60.79 -> 67.07 | Improved |
| compare FontStudio against Adobe Firefly | FontStudio · GenerativeFont benchmark | CLIP-I: 81.74 -> 84.63 | Improved |
| increase noise strength in the shape-adaptive refinement model (ablation) | Shape-adaptive refinement model · Qualitative (letter A) | boundary flexibility vs readability | Mixed |
Later papers that compared with this method
1 later accepted paper reports FontStudio as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- FontCrafter: High-Fidelity Element-Driven Artistic Font Creation with Visual In-Context Generation (CVPR 2026 · better on 6 of 6 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
- Generate Like Experts: Multi-Stage Font Generation by Incorporating Font Transfer Process into Diffusion Models (CVPR 2024 · similarity 0.91)
- FontDiffuser: One-Shot Font Generation via Denoising Diffusion with Multi-Scale Content Aggregation and Style Contrastive Learning (AAAI 2024 · similarity 0.90)
- Diff-Font: Diffusion Model for Robust One-Shot Font Generation (IJCV 2024 · similarity 0.90)
- Awesome Typography: Statistics-Based Text Effects Transfer (CVPR 2017 · similarity 0.89)
- Layout-Agnostic Scene Text Image Synthesis with Diffusion Models (CVPR 2024 · similarity 0.88)
- VecFusion: Vector Font Generation with Diffusion (CVPR 2024 · 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.
