https://github.com/showlab/ringid

https://github.com/showlab/ringid

Last synced: about 2 months ago
JSON representation

• Host: GitHub
• URL: https://github.com/showlab/ringid
• Owner: showlab
• Created: 2024-04-20T16:36:35.000Z (9 months ago)
• Default Branch: main
• Last Pushed: 2024-04-25T06:54:23.000Z (8 months ago)
• Last Synced: 2024-04-28T05:08:06.469Z (8 months ago)
• Size: 15.3 MB
• Stars: 2
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        


RingID 🔍 

RingID: Rethinking Tree-Ring Watermarking for Enhanced Multi-Key Identification


[Hai Ci](https://scholar.google.com/citations?user=GMrjppAAAAAJ&hl=en)*  Pei Yang*  [Yiren Song](https://scholar.google.com/citations?user=L2YS0jgAAAAJ&hl=en&oi=ao)*  [Mike Zheng Shou](https://sites.google.com/view/showlab)

National University of Singapore

[Project Page](https://sites.google.com/view/ringid2?usp=sharing) | [Arxiv](http://arxiv.org/abs/2404.14055)





**RingID** presents a robust diffusion image watermarking approach to imprint multiple keys. It bases on the training-free approach Tree-Ring[1], but significantly enhances in both watermark verification and multi-key identification. 

## Method



RingID identifies the limitations in Tree-Ring's design and suggests a series of approaches for enhanced distinguishability and robustness.

## Dependencies

- PyTorch == 1.13.0

- transformers == 4.23.1

- diffusers == 0.11.1

- datasets

Note: higher diffusers version may not be compatible with the DDIM inversion code.

## Usage

### Verification

```python

python verify.py --run_name verification --online

```

### Identification

```python

python identify.py --run_name identification --online

```

## Citation

```bibtex

@article{ci2024ringid,

  title={RingID: Rethinking Tree-Ring Watermarking for Enhanced Multi-Key Identification},

  author={Ci, Hai and Yang, Pei and Song, Yiren and Shou, Mike Zheng},

  journal={arXiv preprint arXiv:2404.14055},

  year={2024}

}

```

## Acknowledgement

[1] [Tree-ring watermarks: Fingerprints for diffusion images that are invisible and robust](https://github.com/YuxinWenRick/tree-ring-watermark)