Research map / Image generation / Text-to-image diffusion
IGD: Instructional Graphic Design with Multimodal Layer Generation
Yadong Qu, Shancheng Fang, Yuxin Wang, Xiaorui Wang, Zhineng Chen, Hongtao Xie, Yongdong 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 |
|---|---|---|---|
| use IGD parametric text layer rendering instead of diffusion-generated text | IGD · English user inputs | Char-F: 73.83 -> 86.83 | Improved |
| generate editable layers with IGD instead of OpenCOLE | IGD · English user inputs | R_ali: 0.1941 -> 0.0541 | Improved |
| IGD versus DALL-E3 on overall harmony and innovation | IGD · English user inputs (GPT-4o eval) | Harmony: 96.2 vs 88.8 | Hurt |
| add augmented pre-training in stage 1 (ablation) | IGD · GPT-4o eval | Layout and Harmony | Improved |
| convert solid-color and simple pattern images to SVG tokens (ablation) | IGD · training and generation | image quality (qualitative) | Improved |
| use Qwen2.5 instead of Vicuna as the LLM (ablation) | IGD · GPT-4o eval | layout and aesthetic quality | Improved |
| train LLM and SD end-to-end instead of freezing SD (ablation) | IGD · GPT-4o eval | layer consistency | Mixed |
Closest papers by meaning
- Graphic Design with Large Multimodal Model (AAAI 2025 · similarity 0.89)
- Decomposition of Graphic Design with Unified Multimodal Model (ICML 2025 · similarity 0.87)
- DesignDiffusion: High-Quality Text-to-Design Image Generation with Diffusion Models (CVPR 2025 · similarity 0.86)
- Generating Illustrated Instructions (CVPR 2024 · similarity 0.84)
- WeGen: A Unified Model for Interactive Multimodal Generation as We Chat (CVPR 2025 · similarity 0.84)
- LineArt: A Knowledge-guided Training-free High-quality Appearance Transfer for Design Drawing with Diffusion Model (CVPR 2025 · similarity 0.84)
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.
