Research map / Image generation / Text-to-image diffusion
Expressive Text-to-Image Generation with Rich Text
Songwei Ge, Taesung Park, Jun-Yan Zhu, Jia-Bin Huang
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 |
|---|---|---|---|
| build token maps from self-attention spectral clustering plus cross-attention labels instead of cross-attention only (ablation) | Rich-text region-based diffusion · precise color generation (qualitative) | color bleeding into background | Improved |
| inject self-attention maps, residual features and noised plain-text sample (ablation) | Rich-text region-based diffusion · font color and footnote examples (qualitative) | structure fidelity to plain-text generation | Improved |
| use region-based diffusion with font style instead of Prompt-to-Prompt or InstructPix2Pix | Stable Diffusion v1-5 · local style control, 420 prompts | local CLIP similarity | Improved |
| use region-based guidance with RGB font color instead of baselines | Stable Diffusion v1-5 · precise color generation, Common/HTML/RGB categories | L2 distance to target color | Improved |
| reweight tokens with normalized attention instead of unbounded multiplication (Prompt-to-Prompt) | token reweighting with font size · weight 13 for mushrooms (qualitative) | artifacts | Improved |
| apply InstructPix2Pix sequentially for two styles | InstructPix2Pix baseline · local style control | 2nd region CLIP similarity | Hurt |
| use footnotes for region descriptions instead of full lengthy plain text | Stable Diffusion · complex scene generation (qualitative) | detail faithfulness | Improved |
Closest papers by meaning
- DreamText: High Fidelity Scene Text Synthesis (CVPR 2025 · similarity 0.88)
- Brush Your Text: Synthesize Any Scene Text on Images via Diffusion Model (AAAI 2024 · similarity 0.88)
- Layout-Agnostic Scene Text Image Synthesis with Diffusion Models (CVPR 2024 · similarity 0.88)
- TextCenGen: Attention-Guided Text-Centric Background Adaptation for Text-to-Image Generation (ICML 2025 · similarity 0.88)
- R&B: Region and Boundary Aware Zero-shot Grounded Text-to-image Generation (ICLR 2024 · similarity 0.87)
- Rich Human Feedback for Text-to-Image Generation (CVPR 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.
