Tamperproof watermarking of 3D models using hausdorff distance

Abstract

This paper describes a novel algorithm for tamperproof watermarking of 3D models. Fragile watermarking is used to detect any kind of tamper i.e. unauthorized modifications in the model. The best and the simplest way to do this is by inserting a watermark at each and every vertex of the model. This poses as a challenge as insertion of watermark in every vertex can cause perceptible distortion and inserting such a watermark is computationally expensive. The challenge of perceptible distortion is overcome by using a measure that controls perceptible distortional called the hausdorff distance. Thus, the objective of the Genetic Algorithm is to minimize the hausdorff distance between the 2 ring neighbourhood of the original and the watermarked vertex. The other challenge of time complexity is overcome by running the Genetic Algorithm for just 20 generations and causing it to converge prematurely. This significantly reduces the computational cost. The experimental results indicate that the algorithm effectively detects any distortion in model.