傳統的裝飾紋路，如獸面紋和雲紋，是經過工匠藝術家將其對於自然的變化與鳥獸的形態加以抽象化，藉由這些精妙的花紋傳遞豐富的訊息，形成具有象徵意義的紋路。本論文針對傳統中國雲紋進行分析，發現其保有高度的自我相似性，以此特性為基礎，我們發現到L-system也常被運用於此類特性紋路之創建，本論文以此語法為基礎撰寫生長規則，配合parametric L-system參數化的特性，不但可以讓雲紋達到很好的多樣性，同時配合非相片寫實電腦繪圖技術強化風格效果，讓我們的結果更忠於傳統雲紋，更進一步透過具時間序列的向量場資訊影響L-system之成長，產生具東方風格之動態雲紋。 / Traditional decorative patterns, such as animal mask and cloud patterns, are the art heritage from the abstraction of the changing nature and the shape of animals. These decorative patterns are not only exquisite art, but also meaningful symbolic patterns. In this thesis, we analysis the traditional Chinese cloud patterns, and find out the structure of self-similarity in cloud pattern. In this thesis, we design the growth rules base on the grammar of L-system and the parameterization of parametric L-system. We also combine the stylized line drawing to achieve a good variety of cloud patterns, and our results are faithful to the traditional art. Furthermore, we can generate the dynamic oriental cloud patterns with a time-varying vector field to affect the growth rule.

獸面紋又稱做饕餮紋，其盛行於商代至西周早期，但早在夏代青銅器上就已發現這種紋飾，饕餮是古人想像出來的怪獸，進而再融合自然界中各種猛獸的特徵而成。一般形象是以鼻樑為中線，兩側成對稱排列，有大眼和頭上兩個大角，以及一個沒有下唇的大嘴。非相片寫實電腦繪圖技術(Non-photorealistic Rendering)不同於傳統的電腦圖學技術，著重於產生各式各樣不同於光學物理的繪圖效果。最重要的是如何分析各種藝術的結構和技巧，進而加以統整之後，建立系統性的演算法，將其重現於數位空間。本論文便是針對中國古代所流傳下來的獸面紋圖樣的設計創作去做分析，從而發展出一套系統，讓使用者輸入一張任意的動物圖像，將此圖像標示出特徵部位後再與我們收集的獸面紋樣資料庫做簡單的比對和變更，自動的產生繪製出具有中國古代獸面紋圖樣特徵的結果。 / Animal-mask decoration, also known as the taotie pattern, is an art stylization commonly found on bronze vessels from the Shang Dynasty to the early Western Zhou Dynasty. Taotie is an imagined gluttonous monster, which is composited with the characteristics of various wild animals. The general layout of taotie is a symmetrical face with pair of large eyes and horns, and a big mouth without lower lip. Non-photorealistic computer graphics (NPR) technology tries to produce all kinds of effects different from the optical physics, and focuses on analyzing and integrating the principle of the various arts and skills to establish a systematic method to reproduce the stylization. This thesis tries to analyze the structure of the traditional animal-mask decoration, and building a system to transfer the stylization. The User can input an animal photograph and manually select few feature points. Then the proposed algorithm will automatically extract the feature and find the correspondence in the example Database to create the new animal-mask decoration with the feature in the input photograph.

浮雕是雕刻藝術中重要的表現方法，藉由在平板上雕刻出高低落差，傳達出豐富的形狀視覺線索，是介於3D雕塑和2D畫作中間的一種物體外形的表現方式。本論文將針對淺浮雕這類型相對高度較低的浮雕技法，將要表達的3D場景壓縮到接近平面但盡可能保留細節。我們使用適應性直方圖均衡化技術去壓縮高度的動態範圍並盡可能強化細節，且經由降低取樣點數量的技巧加速適應性直方圖均衡化的計算，以利於使用者進行互動性自訂風格化。另外依照場景特徵的流向，增加特殊的刻紋去豐富淺浮雕的風格表現。 / Relief is a sculptural technique to express the shape feature on a flat surface. It is an art medium between 3D sculpture and 2D painting. In this thesis, we focus on bas-relief, which is a relatively low relief to compress the depth of 3D scene to a shallow overall depth and preserve details of the shape. We use the adaptive histogram equalization (AHE) to compress the depth range and enhance details, and accelerate the AHE computation by sample reduction, which is in favor of the user interaction of custom stylization. Furthermore, adding special carving patterns according to feature flows of the scene enriches the stylization of the relief generation.