Third Generation 3 D Watermarking : Applied

With the explosion of multimedia content over Internet, there is a need for copyright protection of digital content. Whether it is music albums swapped over peer to peer networks or video files uploaded over websites such as YouTube.com or 3D models such as Shrek, artists need to protect their ownership of content. 3D Watermarking provides a deterrent to piracy of 3D models by embedding a hidden piece of information in the original content. Watermarking algorithms have a basic requirement that the watermark should be imperceptible to avoid being detected and not cause visible distortion to the viewer. The watermark should also be robust to withstand unintentional attacks. It is also desired that the watermark insertion capacity should be as high as possible to withstand intentional attacks and to allow insertion of multiple or redundant or biometric watermarks. Insertion of high density imperceptible watermark will make it extremely difficult for an attacker to find the watermark and then make substantial changes in the 3D model to remove or overwrite the watermark. However, inserting large amounts of information as watermark can cause distortion. The design of watermarking algorithms involves a trade off between imperceptibility, capacity and robustness. The first generation of 3D watermarking techniques inserted low capacity watermark based on spatial geometry and have poor robustness. The second generation of algorithms explored use of multi-resolution transform to insert the watermark and improve the robustness. This dissertation explores use of computational intelligence techniques to build third generation watermarking algorithms, that insert robust, high density watermarks and go the extra mile in terms of hiding more information than the first and second generation techniques. The focus of this study is to optimize the energy of the watermark and intelligently selecting information pockets in 3D model for watermark insertion and at the same time still maintaining randomness in the process to avoid detection. Use of Fuzzy Logic, Genetic Algorithms, and Artificial Neural Networks are proposed and assessed.