Research map / Image generation / Text-to-image diffusion
De-Rendering Stylized Texts
Wataru Shimoda, Daichi Haraguchi, Seiichi Uchida, Kota Yamaguchi
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 feedback refinement after feedforward parsing | text vectorization · Book cover | PSNR: 29.22 -> 30.02 | Improved |
| add feedback refinement after feedforward parsing | text vectorization · Book cover | L1: 0.15 -> 0.08 | Improved |
| add feedback refinement after feedforward parsing | text vectorization · Advertisement | PSNR: 28.77 -> 29.37 | Improved |
| add feedback refinement after feedforward parsing | text vectorization · Advertisement | L1: 0.16 -> 0.09 | Improved |
| export to external rasterizer instead of differentiable rendering | text vectorization with refinement · Book cover | PSNR: 30.02 -> 29.73 | Hurt |
| export to external rasterizer instead of differentiable rendering | text vectorization with refinement · Advertisement | PSNR: 29.37 -> 29.16 | Hurt |
| refine color parameters (ablation) | refinement · Book cover | PSNR: 29.53 -> 30.02 | Improved |
| refine font parameters (ablation) | refinement · Book cover | PSNR: 29.78 -> 30.02 | Improved |
| refine shadow parameters (ablation) | refinement · Book cover | PSNR: 29.79 -> 30.02 | Improved |
| refine border parameters (ablation) | refinement · Book cover | PSNR: 29.99 -> 30.02 | Improved |
Closest papers by meaning
- SteerVTE: Seamless Video Text Editing with Style and Glyph Control (NeurIPS 2026 · similarity 0.85)
- TextCtrl: Diffusion-based Scene Text Editing with Prior Guidance Control (NeurIPS 2024 · similarity 0.83)
- DreamText: High Fidelity Scene Text Synthesis (CVPR 2025 · similarity 0.83)
- Type-R: Automatically Retouching Typos for Text-to-Image Generation (CVPR 2025 · similarity 0.83)
- GlyphAnchor: Enhancing Visual Text Rendering via Position-Anchored Glyph Priors (NeurIPS 2026 · similarity 0.82)
- AnyText: Multilingual Visual Text Generation and Editing (ICLR 2024 · similarity 0.82)
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.
