【摘要】：隨著數(shù)字化時(shí)代的到來(lái)，傳統(tǒng)動(dòng)畫正在經(jīng)歷著一場(chǎng)從二維走向三維的革命性轉(zhuǎn)變，二維卡通動(dòng)畫形象以其獨(dú)特的視覺形式與審美要求被更多的人認(rèn)可；動(dòng)畫形象也以其視覺傳達(dá)的世界通用“語(yǔ)言”形式成為影響世界文化的因素。如何有效的將三維技術(shù)與二維動(dòng)畫的融合形成新的審美方式已經(jīng)成為了一個(gè)重要的研究課題。 二維卡通動(dòng)畫角色表情往往是多種情感交織的結(jié)果，是高度濃縮了的人性，是現(xiàn)實(shí)生活中人物或非人物的夸張和抽象，因其獨(dú)有的多元化特征很難被確切的劃分為某一特定的表情類別�？v觀現(xiàn)有的面部表情識(shí)別技術(shù)多是采用了參考真人的表情特征，忽略了二維卡通形象面部表情獨(dú)特的視覺形式和審美情感，對(duì)于二維卡通角色表情的識(shí)別和分類方法至今還沒有一個(gè)統(tǒng)一的定義。 現(xiàn)有的主流面部建模技術(shù)與表情動(dòng)畫技術(shù)可以比較精細(xì)的塑造角色面部表情且變形逼真，但是由于實(shí)際操作中這些技術(shù)存在面部模型改造的過程步驟及細(xì)節(jié)調(diào)整復(fù)雜和繁瑣、調(diào)整速度比較慢的缺點(diǎn)；也有因設(shè)置變形參數(shù)繁多，在實(shí)現(xiàn)時(shí)需要反復(fù)比選調(diào)試參數(shù)，耗時(shí)過長(zhǎng)缺乏便捷性的缺點(diǎn)。 隨著開設(shè)動(dòng)漫與影視動(dòng)畫課程的院校日益增多及相關(guān)技術(shù)普及，用優(yōu)化技術(shù)改善教學(xué)過程中的實(shí)現(xiàn)難度，輔助教學(xué)課程學(xué)習(xí)將成為極其有意義的工作。 鑒于此，本文分析二維卡通動(dòng)畫面部表情特征，提出表情識(shí)別思路和分類方法；采用優(yōu)化的三維技術(shù)或方法實(shí)現(xiàn)并調(diào)試給出表情設(shè)置的合理參數(shù)范圍。 本文的研究重點(diǎn)及主要成果有： 1、本文系統(tǒng)介紹了目前主流的角色動(dòng)畫面部表情識(shí)別、面部建模、表情動(dòng)畫等幾方面的實(shí)現(xiàn)技術(shù)，簡(jiǎn)單介紹了現(xiàn)有面部表情技術(shù)的主要特點(diǎn)，并對(duì)角色面部表情識(shí)別與三維技術(shù)實(shí)現(xiàn)方式之間的聯(lián)系和各自的優(yōu)缺點(diǎn)進(jìn)行了詳細(xì)的分析。 2、系統(tǒng)分析二維卡通面部表情特征及對(duì)表情分類進(jìn)行研究，借助符號(hào)學(xué)的相關(guān)理論建立了表情符號(hào)，得出了二維卡通基本表情識(shí)別模型。并用層次化結(jié)構(gòu)方法劃分解決了卡通面部表情的個(gè)體差異性特征。表情符號(hào)模型的構(gòu)建實(shí)現(xiàn)了對(duì)卡通角色反映的情感特征的表達(dá)和還原了其視覺形式效果。 3、采用Pixol建模技術(shù)進(jìn)行表情模型的塑造和局部表情特征調(diào)整，以此改善傳統(tǒng)建模工具進(jìn)行面部模型改造的復(fù)雜過程步驟及繁瑣的細(xì)節(jié)調(diào)整，其快速實(shí)現(xiàn)多元化表情模型的特點(diǎn)，為實(shí)現(xiàn)特定面部表情變形設(shè)置奠定了基礎(chǔ)。 4、提出了一種線控表情綁定融合變形的思路，設(shè)計(jì)采用線控驅(qū)動(dòng)的方法綁定表情的融合變形參數(shù)進(jìn)行調(diào)試表情，以此優(yōu)化變形中表情細(xì)節(jié)設(shè)置的局限，能迅速的對(duì)表情進(jìn)行捕捉，快速獲得特定表情的參數(shù)值的效果，有效實(shí)現(xiàn)了界面互動(dòng)。 5、根據(jù)上述理論與實(shí)現(xiàn)技術(shù)分析，，較為系統(tǒng)的總結(jié)了不同類別表情符號(hào)的特征；對(duì)面部表情符號(hào)進(jìn)行三維模型設(shè)計(jì)，形成特定三維表情模型。通過調(diào)試確定了各類表情符號(hào)的參數(shù)設(shè)置范圍值。為卡通角色動(dòng)畫表情研究的同行提供有價(jià)值的參考。
[Abstract]:With the advent of the digital age, the traditional animation is undergoing a revolutionary transformation from two-dimensional to three-dimensional, and the two-dimensional cartoon image is recognized by more people with its unique visual form and aesthetic requirements; The image of the animation is also the world's universal "language" in its visual communication as a factor that affects the world culture. How to effectively combine the three-dimensional and two-dimensional animation to form a new aesthetic way has become an important research topic. The two-dimensional cartoon character expression is often the result of a variety of emotional interweave. It is a highly concentrated human nature. It is the exaggeration and abstraction of the characters or non-characters in the real life. Because of its unique diversity, it is difficult to be divided into a specific expression class. Don't. Throughout the existing facial expression recognition technology, the facial expression feature of the reference real person is adopted, the unique visual form and the aesthetic feeling of the facial expression of the two-dimensional cartoon image are ignored, and the recognition and classification method for the two-dimensional cartoon character expression has not yet been unified The existing mainstream face modeling technology and the expression animation technology can be used for finely shaping the facial expression of the character and the deformation is vivid, but because the process steps and the detail adjustment of the face model modification in the actual operation are complex and complicated, the adjustment speed is relatively slow, The invention also has the defects that the setting deformation parameters are various, the debugging parameters need to be repeated and selected repeatedly in the implementation, the time consuming is too long and the convenience is poor, With the increasing of the colleges and universities and the popularization of the related technologies, the difficulty in the teaching process is improved by using the optimization technique, and the learning of the auxiliary teaching course will become extremely useful. In view of this, this paper analyzes the facial expression of two-dimensional cartoon animation, puts forward the method of expression recognition and the classification method, and uses the optimized three-dimensional technique or method to realize and debug the expression setting. The reasonable parameter range of this paper. The focus and main results are as follows:1. The present system introduces the current mainstream character animation facial expression recognition, face modeling, expression animation and so on. The main features of facial expression technology, and the relation between the recognition of facial expression and the realization of three-dimensional technology and their respective methods The paper makes a detailed analysis of the advantages and disadvantages of the two-dimensional cartoon facial expression and the study of the classification of the expression. The basic expression recognition model of two-dimensional cartoon is presented, and the method of hierarchical structure is used to solve the problem. The characteristic of the individual difference of the facial expression. The construction of the emoticon model realizes the emotional characteristics of the cartoon character's reflection. and the visual form effect of the traditional modeling tool is improved, and the complex process steps and the complicated detail adjustment of the face model modification of the traditional modeling tool are improved, The method can realize the characteristics of the diversified expression model in a quick way, and is This paper lays a foundation for the setting of specific facial expression.4. It puts forward a method of binding fusion and deformation by wire-by-wire. It is designed to debug the expression by using the method of wire-by-wire method to bind the expression, so as to optimize the expression detail in the deformation. The set limits, can quickly capture the expression, quickly get a specific expression The effect of the parameter value of the parameter value is effectively realized, and the interface interaction is effectively realized. And the three-dimensional model is designed to form a specific three-dimensional expression model. The debugging determines the parameter setting range value of various types of emoticons. It is a cartoon
【學(xué)位授予單位】：上海交通大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2012
【分類號(hào)】：TP391.41

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 劉冬;;《汽車總動(dòng)員》:由汽車表情看“汽車臉譜”的情感化設(shè)計(jì)[J];電影文學(xué);2008年05期

2 王敏;;“超真實(shí)”符號(hào)理論下的動(dòng)畫形象[J];電影文學(xué);2010年07期

3 肖著強(qiáng);李潔;;淺析動(dòng)畫創(chuàng)作中角色造型設(shè)計(jì)的符號(hào)化特征[J];電影評(píng)介;2007年23期

4 鄭玫;;動(dòng)畫片中的表情動(dòng)作與角色塑造——對(duì)《貓和老鼠》動(dòng)畫系列短片的分析[J];福建藝術(shù);2009年04期

5 張滿囤;李智;吳鴻濤;;表情動(dòng)畫技術(shù)和應(yīng)用綜述[J];河北工業(yè)大學(xué)學(xué)報(bào);2007年05期

6 姚俊峰;陳琪;;計(jì)算機(jī)人臉表情動(dòng)畫技術(shù)綜述[J];計(jì)算機(jī)應(yīng)用研究;2008年11期

7 李濤;;動(dòng)畫文化:動(dòng)畫形象的符號(hào)學(xué)研究[J];江西社會(huì)科學(xué);2007年10期

8 許茂強(qiáng);;卡通與卡通形象的運(yùn)用[J];科教文匯(下半月);2006年06期

9 喬鳳天;;符號(hào)學(xué)方法在動(dòng)畫角色研究中的應(yīng)用[J];新聞界;2010年02期

10 王超杰;;初探三維雕刻軟件ZBrush在生物角色建模中的應(yīng)用[J];美術(shù)教育研究;2011年08期

本文編號(hào)：2480612