目前生物模擬方式大多是採用物理計算來模擬出生物的行為，相較於真實的生物，往往會讓人產生不真實的感覺。有鑑於此，我們提出一套視覺學習機制來學習真實生物的行為，使模型可依據影片中學習到的生物行為來模擬動作，產生動畫。在本論文中，先將影像以模板匹配的方式找出生物位置範圍，再利用主動式輪廓(SBGFRLS)分割出生物之輪廓，並基於此輪廓以Delaunay三角化方式計算出二維骨架資訊。我們提出以線性擬合的方式，將兩組不同方向之二維骨架合成，以拓展建立出三維骨架資訊。再將得到的生物骨架資訊及其移動路徑做分析，建立生物的學習資料，讓模型根據此學習資料產生一不受限於影片長度之真實生物的行為模擬動畫。

In this thesis, we present a visual learning mechanism, to learn the realistic motion from real fishes and automatically generate 3D animation. We use two cameras to record the motion of fishes, and tracking the deformable objects by using the Template matching associated with SBGFRLS. After tracking, the skeletons are extracted by the Delaunay triangulation from the contour of creatures. We also proposed a line fitting method to combine the 2D skeletons of two views into the 3D skeleton. Furthermore, the proposed mechanism analyzes the path data and skeleton motions to create the learning data. Finally, one can simulate the continuous animation that not limited to the time length of input videos.

摘 要 i
Abstract ii
目 錄 iii
圖目錄 v
表目錄 vii
壹、 簡介 1
一、 論文概述 1
二、 論文貢獻 2
三、 論文架構 3
貳、 文獻探討 4
一、 影像分割(Segmentation) 5
二、 骨架擷取(Skeleton Extraction) 6
三、 三維重建(3D Reconstruction) 7
四、 動作擷取(Motion Extraction) 7
參、 研究方法 9
一、 影像分割(Segmentation) 11
1、 模板匹配(Template Matching) 11
2、 輪廓擷取(SBGFRLS) 12
二、 動作擷取(Motion Extraction) 15
1、 骨架擷取(Skeleton Extraction) 15
2、 骨架合併(Skeleton Integration) 16
三、 學習機制(Learning Mechanism) 20
1、 資料分析(Data Analysis) 20
2、 建立學習資料表(Create Learning Table) 22
3、 產生動畫(Animation Generation) 23
肆、 系統實作 27
一、 操作介面及環境 28
二、 實作詳細內容 29
伍、 結論與未來目標 42
參考文獻 43

[1] Z. Messaoudi, A. Ouldali, and M. Oussalah, “Tracking objects in video sequence using active contour models and unscented kalman filter,” International Workshop on Systems, Signal Processing and their Applications (WOSSPA), pp. 135-138, 2011.
[2] J. Gall, C. Stoll, E. de Aguiar, C. Theobalt, B. Rosenhahn, and H. Seidel, “Motion capture using joint skeleton tracking and surface estimation,” IEEE Conference on Computer Vision and Pattern Recognition, pp. 1746-1753, 2009.
[3] F. Xu, Y. Liu, C. Stoll, J. Tompkin, G. Bharaj, Q. Dai, H. Seidel, J. Kautz, and C. Theobalt, “Video-based characters: creating new human performances from a multi-view video database,” ACM Transactions on Graphics (TOG), vol. 30, no. 4, pp. 32:1-32:10, 2011.
[4] C. Stoll, J. Gall, E. Aguiar, S. Thrun, and C. Theobalt, “Video-based reconstruction of animatable human characters,” ACM Transactions on Graphics(TOG), vol. 29, no. 6, pp. 139:1-139:10, 2010.
[5] J. Frahm, M. Pollefeys, B. Clipp, D. Gallup, R. Raguram, C. Wu, and C. Zach, “3D reconstruction of architectural scenes from uncalibrated video sequences,” 3D Virtual Reconstruction and Visualization of Complex Architectures (3D-ARCH), vol. 38, no. 5, pp.7, 2009.
[6] S. Zhu, X. Xia, Q. Zhang, and K. Belloulata, “A novel spatio-temporal video object segmentation algorithm,” IEEE International Conference on Industrial Technology, pp. 1-5, 2008.
[7] Y. Zhang, D. Wang, and Y. Hou, “An effective algorithm of automatic video object segmentation based on temporal-spatial information,” International Conference on Computer Science and Education, pp. 1755-1759, 2010.
[8] K. Zhang, L. Zhang, H. Song, and W. Zhou, “Active contours with selective local or global segmentation: a new formulation and level set method,” Journal of Image and Vision Computing, vol. 28, no. 4, pp. 668-676, 2010.
[9] M. Leventon, W. Grimson, and O. Faugeras, “Statistical shape influence in geodesic active contours,” IEEE Computer Society Conference on Computer Vision and Pattern Recognition, pp. 1316-1323, 2000.
[10] S. Dambreville, Y. Rathi, and A. Tannenbaum, “Tracking deformable objects with unscented kalman filtering and geometric active contours,” American Control Conference, pp.2856-2861, 2006.
[11] J. Zou, “A fast skeletonization method,” International Conference on Digital Image Computing: Techniques and Applications, pp. 283-288, 2003.
[12] T. Igarashi, S. Matsuoka, and H. Tanaka, “Teddy: a sketching interface for 3d freeform design,” ACM SIGGRAPH, pp. 409-416, 1999.
[13] J. Wu, G. Zhang, J. Xia, and Z. Cui, “Skeleton extraction of cerebrovascular image based on topology nodes,” International Symposium on Information Processing, pp. 159-162, 2009.
[14] J. Wu, G. Zhang, J. Xia, and Z. Cui, “Gray cerebrovascular image skeleton extraction algorithm using level set model,” Journal of Multimedia, vol. 5, no. 3, pp. 208-215, 2010.
[15] O. Au, C. Tai, H. Chu, D. Cohen-Or, and T. Lee, “Skeleton extraction by mesh contraction,” ACM Transactions on Graphics (TOG), vol. 27, no. 3, pp. 44:1-44:10, 2008.
[16] W. Matusik, C. Buehler, R. Raskar, S. J. Gortler, and L. McMillan, “Image-based visual hulls,” Annual Conference on Computer Graphics and Interactive Techniques, pp. 369-374, 2000.
[17] C. Hernandez, F. Schmitt, and R. Cipolla, “Silhouette coherence for camera calibration under circular motion,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 29, no. 2, pp. 343-349, 2007.
[18] A.Y. Mulayim, U. Yilmaz, and V. Atalay, “Silhouette-based 3-D model reconstruction from multiple images,” IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics, vol. 33, no. 4, pp. 582-591, 2003.
[19] K. Terajima, T. Komuro, and M. Ishikawa, “Fast finger tracking system for in-air typing interface,” Annual CHI Conference on Human Factors in Computer Systems, pp. 3739-3744, 2009.
[20] G. Wen, Z. Wang, S. Xia, and C. Li, “Articulated skeleton fitting,” IEEE International Conference on Computer-Aided Design and Computer Graphics, pp.395-400, 2007.