Surface Modeling In The Game Engine

Surface Modeling has been researched for many years in many fields such as computer graphics (CG) and computer aided geometric design (GAGD). In recent years, with the development of virtual reality and computer games, based on surface reconstruction, surface modeling becomes a hot topic, which is used to build models for objects vividly through computing the point data sampled by three-dimension scanner and using B-spline to fit surface.In this thesis, the problem of surface reconstruction for any surface with arbitrary boundary is resolved and finally a surface reconstruction system based on parameterization is completed. This system is bounded on game engine as the module of engine's functionality. Because the system is used for surface reconstruction, in this thesis we have researched and analyzed the steps for realizing the system and the important algorithms in the steps, and then propose an advanced parameterization algorithm which is the key algorithm in the whole system. The main work in this thesis is as follows:1. B-spline surface reconstruction system is designed based on parameterization. Based on the intensive survey and comprehensive analyses of various approaches to surface reconstruction, the approach based on parameterization is selected as our basal theory. Compared with other approaches, parameterization accords with the basal principle in building models for three-dimension objects. It can decrease the complexity of the operation and improve the efficiency of the system. Moreover, the result of the parameterization can apply in the next steps such as texture mapping and so on. Furthermore, because B-spline surface is flexibility and can express objects in any precision, in stead of traditional triangulation, B-spline surface is selected as our digital models for expressing complex objects. Finally, experiments show that the system is available.2. In this system, linear method is adopted to improve the efficiency of the system. Combined with realizing the well-known algorithms for convex surface efficiently, for concave surface, an improved algorithm is proposed in this thesis. In this algorithm a virtual convex boundary for concave is added first, and then the optimization linear algorithm is used to compute parameterization. The system uses these algorithms to reconstruct any surface with arbitrary boundary, furthermore because all the methods are linear, it can improve the efficiency of the system.