Quantifying Copyright Infringement in AI-Generated Images

This project is focused on evaluating and showcasing the degree of copyright infringement seen in content produced by AI models. It presents four different methods to measure infringements using image similarity metrics the Structural Similarity Index Measure (SSIM), the Information theoretic based Statistic Similarity Measure (ISSM), Root Mean Square Error (RMSE), and Peak Signal-to-Noise Ratio (PSNR). These metrics are utilized to assess images generated by AI models.

Code Directory Structure
/Notebook: This directory houses the pivotal Jupyter notebook, Quantifying Copyright in Generative AI.ipynb, which encapsulates the core analysis. The notebook is a convergence of code, documentation, and visualization, providing a holistic view of the methodology and findings.

/Data: Directory is created when running the colab. Serving as the empirical foundation of the project, this directory is designated for storing datasets or images underpinning the analysis. It is essential that data stewardship principles are adhered to, ensuring compliance with legal and ethical standards.

/Results: Directory is created when running the colab. This directory is the repository for the analytical outputs, encapsulating both quantitative metrics and visual comparisons. It is a testament to the investigative rigor applied, offering tangible artifacts for interpretation and discourse.


Installation Guide
Embarking on the analytical journey necessitates a structured environment setup, ensuring all requisite computational resources are at one's disposal. This segment delineates the process of configuring the analytical environment, detailing the installation of Python, necessary libraries, and the notebook itself. It aims to furnish a seamless setup experience, paving the way for engaging with the analytical narrative encapsulated in the Jupyter notebook.


User Manual
Run the script using a Python interpreter. You may need to pass specific arguments or modify the script to point to your data directory.
Review the output in the /Results directory. This may include textual data on similarity scores or generated images for comparison.

Additional Notes
Modify the paths and directories in the script as necessary to suit your project setup.
Ensure you have the legal right to use any data included in the /Data directory.


The project is designed for research and educational purposes. Ensure that its use complies with local and international copyright laws.