3D Watermarking based on Spectral Decomposition

This repository contains an implementation of the 3D watermarking algorithm proposed by Cayre et al based on Spectral Decomposition. The Original folder contains the implementation of what was exactly described in the paper. The Layered folder contains an alternative to this technique more suited to Additive Manufacturing.

Getting Started

Dependencies

This project uses the following libraries :

• NumPy
• SciPy
• PyMesh

Installation

All the dependencies needed can be installed with the following command:

make init

Usage

The main functions are located inside sample/spectral_decomposition.py :

• insert
• extract

Running the tests

There is no unit testing implemented at this time. To perform tests, simply run make tests-original or make tests-layered. This command will first insert a watermark then read it back and print the number of errors found. The visual deformations introduced by the watermark can be observed by opening the tests/watermarked_models/out.obj file.

Author

• Antoine Prieëls - aprieels

See also the list of contributors who participated in this project.

License

This project is licensed under the MIT License - see the LICENSE.md file for details

Acknowledgments

• Benoît Macq and Patrice Rondao-Alface for their help during the development