Geometry Compression for 3D Polygonal Models using a Neural Network

Three dimensional models are commonly used in computer graphics and 3D modeling characters in animation movies and games. 3D objects are more complex to handle than other multimedia data due to the fact that various representations exist for the same object, yielding a number of difficulties, among of which are the distinct sources of 3D data. Research work in the field of three dimensional environments is represented by a broad spectrum of applications. In this paper we restrict ourselves only on how to do compression using a neural network in order to minimize the size of 3D models for making transmission over networks much faster. The main objective behind this compression is to simplify the 3D model and make handling the large size of 3d objects much easier for other processes. Even the process of rendering, digital watermarking, etc., will be faster and more efficient.