https://github.com/byliutao/1Prompt1Story

🔥ICLR 2025 (Spotlight) One-Prompt-One-Story: Free-Lunch Consistent Text-to-Image Generation Using a Single Prompt
https://github.com/byliutao/1Prompt1Story

diffusion diffusion-model diffusion-models storytelling text-to-image

Last synced: 5 days ago
JSON representation

🔥ICLR 2025 (Spotlight) One-Prompt-One-Story: Free-Lunch Consistent Text-to-Image Generation Using a Single Prompt

• Host: GitHub
• URL: https://github.com/byliutao/1Prompt1Story
• Owner: byliutao
• License: mit
• Created: 2024-11-18T12:41:59.000Z (11 months ago)
• Default Branch: main
• Last Pushed: 2025-05-31T07:34:06.000Z (5 months ago)
• Last Synced: 2025-05-31T19:07:15.938Z (5 months ago)
• Topics: diffusion, diffusion-model, diffusion-models, storytelling, text-to-image
• Language: Python
• Homepage: https://byliutao.github.io/1Prompt1Story.github.io/
• Size: 29.9 MB
• Stars: 265
• Watchers: 1
• Forks: 32
• Open Issues: 10
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

• awesome-diffusion-categorized - [Code

README

          


  🔥ICLR 2025 (Spotlight) One-Prompt-One-Story: Free-Lunch Consistent Text-to-Image Generation Using a Single Prompt

  








    

 



    

 



    

 





  Key Features •

  How To Use •

  License •

  Citation •

  Visualization 



[![Demo](./resource/demo.png)](./resource/demo.png)

## Key Features

* Consistent Image Generation Code: main.py

* Gradio Code: app.py 

* Benchmark Generation Code: resource/gen_benchmark.py

* Consistory+ Benchmark: [link](./resource/consistory+.yaml)

* Online Gradio demo: [link](http://59.110.93.62:7089/)

* 🚩[Chinese Version Paper (new)](https://www.overleaf.com/read/pxyhqpsccfjk#20922a)

* 🚩[Jittor Version (new)](https://github.com/byliutao/1Prompt1Story/tree/main/jittor_version)

## How To Use

```bash

# Clone this repository

$ git clone https://github.com/byliutao/1Prompt1Story

# Go into the repository

$ cd 1Prompt1Story

### Install dependencies ###

$ conda create --name 1p1s python=3.10

$ conda activate 1p1s

# choose the right cuda version of your device

$ conda install pytorch torchvision torchaudio pytorch-cuda=12.1 -c pytorch -c nvidia 

$ pip install transformers==4.46.3  # or: conda install conda-forge::transformers 

$ conda install -c conda-forge diffusers

$ pip install opencv-python scipy gradio==4.44.1 sympy==1.13.1

### Install dependencies ENDs ###

# Run infer code

$ python main.py

# Run gradio code

$ python app.py

# Run benchmark generation code

$ python -m resource.gen_benchmark --save_dir ./result/benchmark --benchmark_path ./resource/consistory+.yaml

```

## License

This project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.

## Citation

If our work assists your research, feel free to give us a star ⭐ or cite us using:

```

@inproceedings{

liu2025onepromptonestory,

title={One-Prompt-One-Story: Free-Lunch Consistent Text-to-Image Generation Using a Single Prompt},

author={Tao Liu and Kai Wang and Senmao Li and Joost van de Weijer and Fhad Khan and Shiqi Yang and Yaxing Wang and Jian Yang and Mingming Cheng},

booktitle={The Thirteenth International Conference on Learning Representations},

year={2025},

url={https://openreview.net/forum?id=cD1kl2QKv1}

}

```

## Visualization

### Multi-Subject Generation

  

    

  

### Comparison with Other Models

  

    

  

### Long Story Image Generation