svjack/bizgen_infographic_bbox

BizGen: Advancing Article-level Visual Text Rendering for Infographics Generation (Glyph-ByT5-v3) (BBox Output, As layout Input)

Dataset Overview

This dataset supports research in article-level visual text rendering for business content generation, including infographics and presentation slides with ultra-dense layouts.

Key Features

📊 Layout Capabilities

• Supports ultra-dense layouts with 50+ layers • Handles article-level descriptive prompts (>1000 tokens) • Generates high-resolution outputs up to 2240×896 pixels

🌍 Multilingual Support

• Provides visual text rendering in 10 languages • Maintains high spelling accuracy across languages

🎨 Generation Flexibility

• Enables layer-wise detail refinement through layout-conditional CFG • Supports diverse business content generation (infographics, slides, etc.)

Usage

1. Download the dataset from Hugging Face
2. Refer to their paper for detailed methodology
3. Visit their project page for visual examples

Citation

If you use this dataset in your research, please cite their work:

@article{bizgen2024,
  title={BizGen: Advancing Article-level Visual Text Rendering for Infographics Generation},
  author={Author List},
  journal={arXiv preprint arXiv:2503.20672},
  year={2024}
}

License

This dataset is released for academic research purposes only. Commercial use requires separate permission.--- # For reference on dataset card metadata, see the spec: https://github.com/huggingface/hub-docs/blob/main/datasetcard.md?plain=1 # Doc / guide: https://huggingface.co/docs/hub/datasets-cards {}