附着贴图的3维角色模型融合方法

Method of model fusion for 3D-character with maps

现有的3维模型融合均是立足于模型的几何网格信息,需要大量的顶点来提升融合效果。然而在目前的游戏与动画中,角色3维模型一般使用低精度模型,而以贴图来表现模型细节。因此针对低模角色网格顶点稀疏以及贴图特征,提出简单调和映射和均平面的思想简化网格连接,提高网格融合效率;其次采用基于仿射变换的贴图合成方法完成网格模型融合后的新贴图生成问题,进而完成角色模型的无缝融合。针对本文的方法以及3维模型设计制作的业务流程,建立人机交互的3维模型融合应用系统,实现高效率的3维模型融合与贴图合成。实验结果表明,该方法能够快速有效地完成动画和游戏中角色模型的融合,形成逼真度较高的新角色,为动画和游戏应用增添乐趣。

The existing methods of mesh fusion for 3D models are based on geometric mesh, which needs sufficient number of vertices to enhance the purpose, to connect components. However, the 3D characters in current games and animations generally use low poly model, which mainly uses a variety of maps to show details. To befit the situation of the low-model characters＇ sparse vertices and extensive use of maps, we propose a simple method of harmonic mapping and averaged-plane to simplify mesh connection process and improve the efficiency. Thereafter, map synthesis based on affine transformation is applied to solve the problem of generation of the new map. Therefore, the new seamless fused model is acquired. Based on methods used in this paper and operational processes of model design and production, a Human-Computer interactive fusion application is developed to implement the efficient 3D model fusion and map synthesis operations. Many experimental results have shown that this approach is fast and effective enough to accomplish the fusion for 3D-character models in games and animations, forms a new character model that is more vivid, and thereby makes more interesting applications.