近幾年來,因顯示卡繪圖晶片(Graphic Process Unit、GPU)的進步對於複雜模型的繪製速度提升了不少。但多邊形繪製系統面對單一複雜模型(即擁有大量多邊形構成的模型) 的繪製處理過程中,會發現多邊形投影到螢幕上可能遠小於一個像素(pixel)的大小,但卻花費了許多的時間在shading的運算。 為了提升系統繪製效能,本研究採用Point-Based Rendering System(PBR)來繪製複雜模型。由於模型是以點來繪製,故在模型的shading運算效能上較多邊形繪製快。且使用PBR於模型動作時,可以免於多邊形繪製系統因動作所遇到模型表面細分(subdivision)及重疊交錯(self intersections)的問題,如此一來就可以提升模型在real-time動作呈現的效能。
In recent years, due to computer hardware and display card’s progress, there are great developments and achievements in 3D computer graphics. It is time-consuming for Polygon-Based Rendering System to render the complex model. Because overhand for triangle setup increases (ex: rasterization), it is amortized over fewer and fewer pixels, leading to a higher per pixel rendering cost. In order to promote the effort of system, we use Point-Based Rendering System to render the complex model in our research. Because the Point-Based Model is composed of many points, the shading computation can be promoted without triangle setup. When we use Point-Based Rendering System for the animating model, we can avoid subdivision and self-intersection on model’s surface for Polygon-Based Rendering System. Then, we can save the rendering time to solve two problems. So, we can render the animating model in real-time.