透過您的圖書館登入
IP:3.143.218.146
  • 學位論文

資料採礦應用於人體動作質地分析-以LMA為基礎

Applying Data Mining Technique to Effort of Body Movement Analysis–Based on LMA

指導教授 : 方鄒昭聰
若您是本文的作者,可授權文章由華藝線上圖書館中協助推廣。

摘要


在現今數位科技時代,資料採礦技術被普遍使用於商業環境,為近年來發燒的話題-商業智慧(Business Intelligence)非常重要的部份,多半用於尋找目標顧客群、信用風險預測…等支援商業活動之決策,本研究不同於以往,將此一技術運用在人體動作質地的分析上,嘗試找出人體動作的”本質”,而非目前數位化動作記錄中單純記錄動作的”外型”而已。在動作分析學理部份 ,我們採用由德國舞蹈學家Rudolf Laban所發明的”拉邦動作分析(Laban Movement Analysis, LMA)”來作為研究基礎,它是一套有系統且完整的動作分析理論,其中關於Effort(質地)的論述是該學說中很重要的一環,但它是對動作描述性詞彙的集合,並沒有一個可量化的方法來對動作進行數位分析(Chi, D., Costa, M., Zhao, L., & Badler, N., 2000),故我們透過計算運動學中的8種參數- 位移、累積位移、速度、加速度、角度變化量、累積角度變化量、角速度、角加速度,來獲得動作中人體肢段及關節變化的量化數值。 本研究主要目的為建立人體動作數位化分析的採礦模型,利用決策樹演算法找出動作”質地(Efforts)”與由人體肢段/關節及運動學參數組成的動作 ”因子(Factors)”間的關係,並建置動作質地資料庫,以做為3D遊戲產業、運動指導系統、動畫製作等數位化應用。

並列摘要


In the digital technology age nowadays¸ data mining techniques had been applied to business environment widely, and it’s a significant part of Business Intelligence which was used to support decision making of business activities, for example, target customer group exploring, credit risk prediction, etc. This study is applying these techniques to body movement analysis, which was trying to find out the essence of movement rather than just record the “outer shape” of movements. We adopt Laban Movement Analysis (LMA) advanced by Rudolf Laban as the fundamental theory of this research. It’s a systematic and integrated movement analysis theory. The comment of “Effort” is an important part of the theory, but it’s only the aggregation of descriptive sentences for movements, and there is no quantifiable method for digital analyzing (Chi, D., Costa, M., Zhao, L., & Badler, N., 2000). We acquire the variation quantitative value of body parts and joints in movements through the calculation of 8 kinematic parameters – Displacement, Cumulated Displacement, Velocity, Acceleration, Angle Variation, Cumulated Angle Variation, Angle Velocity, Angle Acceleration. The main purpose of the research is to establish a data mining model for digital analysis method of body movement, by using decision-tree approaches to find out the relationship between “Movement Efforts” and “Movement Factors”(Composition of body parts/joints and kinematic parameters), and construct MEDB (Movement Effort DataBase) for the digital application to 3D Entertainment Industry, Motion Guiding System and Animation production.

參考文獻


8.陳五洲 & 黃彥慈 (2007),拉邦動作分析論,大專體育,第88期,pp.169-175。
1.方鄒昭聰 & 許世賢 (2008),基於拉邦動作分析的人體動作質地分析與診斷模型之建立,2008數位科技與創新管理研討會,華梵大學,台北。
5.Fangtsou, C. T., Yang, T. C. (2008). About Body Movement Factors by Using K-Means Algorithm. Proceedings of the e-CASE 2008 International Joint Conference on Advances in e-Commerce.
7.張中煖 (2002),現代舞教學源流初探,藝術評論,第13期,pp.273-284。
1.Badler, N., Chi, D., & Chopra, S. (1999). Virtual human animation based on movement observation and cognitive behavior models. Proceedings of the Computer Animation Conference, (pp.128-137).

被引用紀錄


楊迪強(2008)。運用K-Means演算法探討人體動作因子之組成-以LMA為基礎〔碩士論文,國立臺北大學〕。華藝線上圖書館。https://www.airitilibrary.com/Article/Detail?DocID=U0023-0308200800252800
蘇昱瑋(2009)。以LMA為基之人體動作因子權值分析〔碩士論文,國立臺北大學〕。華藝線上圖書館。https://www.airitilibrary.com/Article/Detail?DocID=U0023-2408200911152700
黃瑋珊(2009)。基於拉邦動作分析理論之數位化動作質地診斷模型〔碩士論文,國立臺北大學〕。華藝線上圖書館。https://www.airitilibrary.com/Article/Detail?DocID=U0023-1908200911034300
林婉婷(2012)。火警初期派遣決策支援系統反饋機制研究-以新北市為例〔碩士論文,國立臺北大學〕。華藝線上圖書館。https://www.airitilibrary.com/Article/Detail?DocID=U0023-1007201203443100

延伸閱讀