Research On Hair Modeling And Real-Time Rendering

Hair rendering has always been one of the most important and difficult problems in 3D rendering.The quality of hair rendering directly affects the reality of virtual human in virtual reality technology.In the current hair modeling and rendering solutions,there are still many problems,such as inability to render real-time,inefficiency,high resource consumption and so on.Therefore,this thesis studies the solution of hair modeling and rendering.The main research work is as follows:1.In hair modeling,this thesis improves the traditional hair spring-mass model.By adding additional constraints,the problem of excessive bending and shaking of hair silk in the physical simulation process is solved.Then,In view of the simulation process,the force acting on the particle and its corresponding dynamic constraints are discussed in detail.Then,this thesis discusses the basic method of solving the equation dynamics,and introduces a new wind field model,which improves the reality of hair simulation in the wind field.Finally,the bounding sphere is introduced to provide collision detection solutions for hair simulation.2.In hair rendering,this thesis first briefly introduces the new generation of programmable OpenGL rendering pipeline.Then the ArcBall algorithm for manipulating three-dimensional objects in two-dimensional screen is introduced.After that,the basic illumination models of hair rendering are discussed,including traditional Kajiya-Kay model and Marschner model.Aiming at the problem that the number of interpolation is difficult to determine in traditional hair simulation,a triangular viewport deployment technology based on billboards is proposed.In addition,the basic methods of hairpiece subdivision,interpolation and anti-aliasing are introduced.Finally,based on the above research,a virtual hair simulation system is developed by using Qt and OpenGL.Experiments show that the system can simulate the hair of virtual characters very well,has a good sense of hair reality,and can meet the dynamic physical simulation.At the same time,this method occupies very little computing resources and can meet the real-time requirements very well.