跳到主要內容

臺灣博碩士論文加值系統

(44.200.86.95) 您好!臺灣時間:2024/05/18 11:12
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果 :::

詳目顯示

: 
twitterline
研究生:梁冬昇
研究生(外文):LIANG, DONG-SHENG
論文名稱:基於3D點雲重建之自動化身型建模與衣服打板方法
論文名稱(外文):An Automatic Body Modeling and Cloth Patterning Method based on 3D Point Cloud Reconstruction
指導教授:張厥煒張厥煒引用關係
指導教授(外文):CHANG, CHUEH-WEI
口試委員:奚正寧楊士萱張厥煒
口試日期:2019-06-14
學位類別:碩士
校院名稱:國立臺北科技大學
系所名稱:資訊工程系
學門:工程學門
學類:電資工程學類
論文種類:學術論文
論文出版年:2019
畢業學年度:107
語文別:中文
論文頁數:63
中文關鍵詞:點雲函式庫人體模型人體量度點雲衣服打板
外文關鍵詞:Point Cloud Library3D Human Body Modeling3D Human Body MeasurePoint CloudCloth Patterning
相關次數:
  • 被引用被引用:0
  • 點閱點閱:393
  • 評分評分:
  • 下載下載:75
  • 收藏至我的研究室書目清單書目收藏:0
在三維人體模型於現代醫學、遊戲角色、服飾業等領域應用普及化之前,以往只能利用二維影像作物件的判斷。但使用二維影像建立三維模型時因資訊不足而使三維模型不夠精確。在加入點雲空間元素後,以往僅利用二維資訊建模而缺少的資訊便能分析出來。本研究將使用點雲建立三維人體模型尋找人體特徵點,並將特徵點抽取轉化爲服裝業版型設計圖,讓製作服裝版型有一套自動化的流程,節省人工製圖以及量測以加速產品開發。
本研究將自動化系統分為人體建模、人體量度與板型製作自動化三部分。人體建模使用深度攝影機,以3D掃描技術獲取人體點雲數據後,將數據以 Obj與Ply等點雲格式儲存。第二部分,透過點雲函式庫(PCL)將原始資料進行濾波、正規化等將點雲資訊前處理。之後依照人體特徵點將計算後的點雲座標由像素轉換為英吋作為人體數值。
第三部分為版型製作自動化,使用人體建模時獲得的數值經由電腦視覺庫(OpenCV) 轉換並帶入版型製作書之公式以製作設計圖。版型設計圖完成後提供使用者微調版型參數。並將版型輸出圖後處理為適合個人工作室印表機尺寸後印出,即完成版型自動化。

Before 3D human model has been widely in industries such as medical, gaming, and fashion industry. The industries are only using 2D images for 3D modeling. But human is not like mechanical parts, it is hard to modeling human for information only from 2D drawings. Instead of only using 2D images to build 3D model, adding point cloud space element can fulfill the missed information for 3D modeling. This research is using point cloud to build 3D human model, and use the 3D human model to get the design measurements for cloth pattern design. By automation the fitting, measuring and drawing process, it expedite the product development timing and save the labor cost.
The automation system in this research has three parts, 3D human modeling, 3D human measure and automatic patterning. To build a human model, this research use depth camera 3D scanning technology to get the point cloud data. Then import the point cloud data as OBJ and PLY format. Second part, restructure the point cloud data by filtering, normalization through Point Cloud Library(PCL). Last step of human modeling is convert data from voxel to inch as fitting measurement.
Third part of the system is automated the patterning process. Covert the measurement data from 3D human modeling through OpenCV and use the converted data as design parameters to create pattering layout follow the design book equation. Finally, after fine tune the parameter data by designer, the pattering sketch will be after-processing to be printable in A4 size.

摘 要 i
ABSTRACT ii
誌謝 iv
目錄 v
表目錄 vii
圖目錄 viii
第一章 緒論 1
1.1 研究動機 1
1.2 研究目的 2
1.3 研究範圍與限制 2
1.4 論文架構 4
第二章 相關研究與文獻探討 5
2.1 三維物體建模 5
2.2 成衣打版技法 6
2.3 三維模型格式 8
2.4 點雲(Point Cloud) 8
2.5 三維人體相關文獻 9
第三章 系統架構與流程 11
3.1 設計理念 11
3.2 系統架構 11
3.2.1 人體掃描模組 12
3.2.2 人體量度模組 13
3.2.3 打版繪製模組 13
3.3 系統流程 13
第四章 三維人體建立 16
4.1 定界框設定 17
4.2 即時3D建模 20
第五章 三維人體量度 23
5.1 正規化 25
5.1.1 地板判斷 26
5.1.2 地板正規化 28
5.1.3 人體正規化 29
5.2 胸圍尺寸量度 31
5.3 肩寬尺寸量度 35
第六章 版型繪圖模組 38
6.1 設定初始畫板 39
6.2 繪製上衣原型 40
6.3 印出打版原型 42
第七章 實驗結果 44
7.1 系統開發環境 44
7.2 實驗運作之硬體介紹 45
7.3 三維人體建模 46
7.4 三維人體量度 49
7.5 版型製作 54
7.6 成品展示 57
第八章 結論與未來展望 59
8.1 結論 59
8.2 未來展望 59
參考文獻 61


[1] 紡拓會,台灣紡織工業概況,取自https://www.textiles.org.tw/TTF/main/content/ContentDesc.aspx?menu_id=95
[2] Amazon Wall Street Journal, "Amazon Wants to Know Your Waistline,"Retrieved from https://www.wsj.com/articles/amazon-studies-body-sizes-to-get-that-perfect-clothing-fit-1525355115, 2018.
[3] Microsoft, "Kinect for Windows", Retrieved from https://www.kinectforwindows.org, 2014.
[4] Mur-Artal, Raul, Jose Maria Martinez Montiel, and Juan D. Tardos, "ORB-SLAM: a versatile and accurate monocular SLAM system," IEEE transactions on robotics 31.5, 2015, pp.1147-1163.
[5] ZenFone AR, Retrieved from https://www.asus.com/tw/Phone/ZenFone-AR-ZS571KL/
[6] Point Cloud, Retrieved from https://en.wikipedia.org/wiki/Point_cloud
[7] Microsoft Kinect Fusion, Retrieved from https://www.microsoft.com/en-us/research/publication/kinectfusion-real-time-3D-reconstruction-and-interaction-using-a-moving-depth-camera/
[8] Rusu, Radu B., and S. Cousins, "Point cloud library (pcl)," 2011 IEEE international conference on robotics and automation, 2011.
[9] Bradski, Gary, and Adrian Kaehler, Learning OpenCV: Computer vision with the OpenCV library, " OReilly Media, Inc.", 2008.
[10] 翁麗明、蘇惠玲,男裝&童裝成衣打版技法,臺北:北星圖書事業,第1-17頁,2005。
[11] Structure from motion point cloud calculating, Retrieved from https://openmvg.readthedocs.io/en/latest/_images/structureFromMotion.png
[12] 打版名詞介紹,取自 http://dorafundiy.blogspot.com/2016/07/symbol-pattern-paper.html
[13] Lin, Yueh-Ling, and Mao-Jiun J. Wang, "Constructing 3D human model from 2D images," 2010 IEEE 17Th International Conference on Industrial Engineering and Engineering Management, IEEE, 2010.
[14] Yuan, Renqi, and Zengbo Xu, "Fast human body 3D data acquisition based on multi body sensor," 2016 IEEE Advanced Information Management, Communicates, Electronic and Automation Control Conference (IMCEC), IEEE, 2016.
[15] 丁瑜,服飾設計自動化-自動版型的產生,碩士論文,成功大學機械工程學系碩博士班,台南,2004。
[16] QT, Retrieved from https://www.qt.io/
[17] Lempitsky, Victor S., et al, "Image segmentation with a bounding box prior," ICCV, Vol. 76, 2009.
[18] Izadi, Shahram, et al, "KinectFusion: real-time 3D reconstruction and interaction using a moving depth camera," Proceedings of the 24th annual ACM symposium on User interface software and technology, ACM, 2011.
[19] Rusinkiewicz, Szymon, and Marc Levoy, "Efficient variants of the ICP algorithm," 3dim, Vol. 1, 2001.
[20] Drebin, Robert A., Loren Carpenter, and Pat Hanrahan, "Volume rendering," ACM Siggraph Computer Graphics, Vol. 22, No. 4, ACM, 1988.
[21] Drouin, Simon, and D. Louis Collins, "PRISM: An open source framework for the interactive design of GPU volume rendering shaders," PloS one 13.3 , 2018.
[22] Fischler, Martin A., and Robert C. Bolles, "Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography," Communications of the ACM 24.6 , 1981, pp. 381-395.
[23] PassThrough Filter. Retrieved from http://docs.pointclouds.org/trunk/classPCL_1_1_pass_through.html
[24] 三維平面方程式,取自https://imagingsolution.net/math/plane_equation/
[25] Dai, Jian S, "Euler–Rodrigues formula variations, quaternion conjugation and intrinsic connections," Mechanism and Machine Theory 92 , 2015, pp. 144-152.
[26] Weisstein, Eric W, "Rotation Matrix," From MathWorld--A Wolfram Web Resource, http://mathworld.wolfram.com/RotationMatrix.html
[27] Kanungo, Tapas, et al, "An efficient k-means clustering algorithm: Analysis and implementation," IEEE Transactions on Pattern Analysis & Machine Intelligence 7 , 2002, pp. 881-892.
[28] Unser, Michael, Akram Aldroubi, and Murray Eden, "B-spline signal processing. I. Theory," IEEE transactions on signal processing, 1993, pp. 821-833.
[29] Bezier Curve, Retrieved from https://en.wikipedia.org/wiki/B%C3%A9zier_curve
[30] Plotter, Retrieved from https://en.wikipedia.org/wiki/Plotter
[31] Cutting Plotter, Retrieved from https://www.amazon.com/Cutter-Plotter-Cutting-Signmaster-Software/dp/B00WTUHSGI?language=en_US
[32]Microsoft Kinect v2, Retrieved from https://blogs.msdn.microsoft.com/kinectforwindows/2014/06/05/pre-order-your-kinect-for-windows-v2-sensor-starting-today/

QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top