Perceptual 3D model hashing using key-dependent shape feature

Abstract

With the rapid growth of three-dimensional (3D) content, perceptual 3D model hashing will become a solution for the authentication, reliability, and copy detection of 3D content and will continue to be an important aspect of multimedia security in the future. However, perceptual 3D model hashing has not been used as widely as perceptual image or video hashing. In this study, a robust and secure perceptual 3D model hashing function is developed based on a key-dependent shape feature. The main objectives of our hashing function are to exhibit robustness against content-preserved attacks and to enable blind-detection without the use of preprocessing techniques for these types of attacks. In order to achieve these objectives, our hashing projects all of the vertices to the shape coordinates of the shape spectrum descriptor and the curvedness, and then, it segments the shape coordinates into irregular cells and computes the shape features of the cells using a permutation key and a random key. A perceptual hash is generated by binarizing the shape features. Experimental results confirm that the proposed hashing scheme shows robustness against geometrical and topological attacks and provides a unique and secure hash for each model and key.