Research map / Image generation / Text-to-image diffusion
DS-Fusion: Artistic Typography via Discriminated and Stylized Diffusion
Maham Tanveer, Yizhi Wang, Ali Mahdavi-Amiri, Hao Zhang
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 discriminator on glyph feature maps to latent diffusion fine-tuning | DS-Fusion (single-font) · Artistic typography OCR | OCR accuracy: 7.0 -> 36.4 (vs SD) | Improved |
| single-font mode versus the others on style incorporation | DS-Fusion (single-font) · Artistic typography CLIP score | CLIP score: 23.1 -> 21.1 (vs SD) | Hurt |
| use multi-font training instead of single-font (ablation) | DS-Fusion · Artistic typography | OCR accuracy / CLIP score: OCR 36.4 -> 5.1; CLIP 21.1 -> 24.1 | Mixed |
| increase discriminator weight lambda (ablation) | DS-Fusion · Mermaid + G(M) qualitative | glyph shape fidelity | Improved |
| increase number of generated style images beyond 25 (ablation) | DS-Fusion · Qualitative | generation quality and diversity | No clear effect |
| increase number of style images from very low to 25 (ablation) | DS-Fusion · Qualitative | generation quality and diversity | Improved |
| replace BERT text encoder with CLIP (ablation) | DS-Fusion · Qualitative | generality | No clear effect |
Later papers that compared with this method
1 later accepted paper reports DS-Fusion as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- VecDesigner: Exploring Visual Guidance and Structural Consistency for Semantic Typography (ICML 2026 · better on 3 of 3 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
- FontDiffuser: One-Shot Font Generation via Denoising Diffusion with Multi-Scale Content Aggregation and Style Contrastive Learning (AAAI 2024 · similarity 0.89)
- FontStudio: Shape-Adaptive Diffusion Model for Coherent and Consistent Font Effect Generation (ECCV 2024 · similarity 0.89)
- MetaDesigner: Advancing Artistic Typography through AI-Driven, User-Centric, and Multilingual WordArt Synthesis (ICLR 2025 · similarity 0.87)
- Layout-Agnostic Scene Text Image Synthesis with Diffusion Models (CVPR 2024 · similarity 0.87)
- DesignDiffusion: High-Quality Text-to-Design Image Generation with Diffusion Models (CVPR 2025 · similarity 0.87)
- RealCompo: Balancing Realism and Compositionality Improves Text-to-Image Diffusion Models (NeurIPS 2024 · 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.
