以感知為基礎的非寫實描繪架構：模擬映像派畫作

Abstract

呈現畫家風格描繪演算法已經吸引很多學者的注意，不過，目前的演算法既不是有系統的也不易評估。在這本論文中，我們提出了一種以感知為基礎用於呈現畫家風格描繪演算法的非寫實描繪架構。這個架構由三個子系統組成︰原始物件映射系統、描繪系統及畫作實作系統。原始物件映射系統負責各種場景物件(二維區域，一維曲線和零維點)之間的映射；描繪系統負責將各種場景物件用畫家風格來描繪；畫作實作系統給予真正的畫作物理模擬。每個子系統對映到C++中的一個抽象類別，使用者可以繼承這些類別來實作出自訂的風格。使用這個框架，呈現不同畫家風格的描繪可以在同一個系統中被實現。另外，這個架構的劃分是基於真實畫家的作畫過程，所以其結果的產生更具有美學價值。在本論文中，我們藉由運用此架構來合成印象派畫作以顯示它的效力。

Painterly rendering has drawn many experts’ attention. However, the state of the art is neither systematic nor evaluative. In this thesis, a perceptually-based NPR framework for painterly rendering is presented. The framework consists of three sub-systems: primitive mapping system, rendering system and mark system. The primitive mapping system responds for mapping between various scene primitives (2D region, 1D line and 0D point); the rendering system takes charge of synthesizing different painterly styles based on various scene primitives; the mark system gives the actual physical implementation of the strokes generated by the rendering system. Each sub-system is written as an abstract class in C++ language which can be inherited by user to do some custom control. Using this framework, different painterly algorithm can be implemented in this system. Besides, the division of this framework is based on the actual process of painters which means the results generated can be more evaluative in esthetics aspect. In this thesis, Impressionist paintings are synthesized by this framework to show its effectiveness.