臺灣博碩士論文加值系統

　　建立一個真實的三維人臉模型在電腦視覺這個領域裡面一直以來都是一項挑戰。我們發展了一套系統，讓我們可以容易又有效率的重建出三維人臉模型。我們這個系統使用單一個攝影機(普通的攝影機或是網路攝影機都可以)來擷取人臉在攝影機前面做轉動動作的影片之後，選取一些特徵點，利用光學流動(Optical flow)來做追蹤，找到在每一個畫面裡面每一個特徵點的對應點之後，使用正交投影模型(Orthogonal projection model)的因數分解法(factorization)找出三維的人臉模型以及每一個畫面的轉動向量。之後，我們將使用直角投影模型重建出來的結果，改進其結果使其逼近真實世界的投影方法：透視投影法(Perspective projection)，讓重建的結果更好。我們的系統同時也解決了某種程度上的特徵點被遮蔽的問題。最後，因為選取的特徵點數目不夠多，導致結果看起來有稜有角的，所以我們發展了一個方法，在空間中找出一些額外的點來讓重建出來的結果看起來比較平滑。

　Generating realistic three-dimensional (3D) human face models has been a persistent challenge in computer vision. We developed a system that can easily and efficiently reconstruct 3D faces. The system takes a video sequence of the face having a pan rotation from one single regular camera. Based on the corresponding points tracked by optical flow, we can reconstruct the 3D face using the factorization approach in an orthogonal projection model. Then we improve the orthogonal model to approximate the perspective model to get a more reliable result. The proposed system also solves the missing point problem. Finally our smoothing method can make the 3D face look smoother and more realistic by interpolating additional vertices.

List of Figures　　　　　　　　　　　　　　　　　vi
List of Tables　　　　　　　　　　　　　　　　　　vii
Chapter 1. INTRODUCTION　　　　　　　　　　　　　1
1.1. Motivation and Goal　　　　　　　　　　　1
1.2. Previous work　　　　　　　　　　　　　　2
Chapter 2. SYSTEM DESCRIPTION　　　　　　　　　　　4
2.1. Data Collection and Finding Corresponding Points from Video Sequence 　　　　　　　　　　　　　　　　　　　　　5
2.2. Factorization Approach Using the Orthogonal Model 6
2.3. Approximating to the Perspective Model　　8
2.4. Reconstruction with Missing Points　　　　13
2.5. Smoothing the 3D Face Surface　　　　　　15
Chapter 3. EXPERIMENTAL RESULTS　　　　　　　　　　19
Chapter 4. CONCLUSTION　　　　　　　　　　　　　　33
REFERENCES　　　　　　　　　　　　　　　　　　　　35

[1]T. Akimoto, Y. Suenaga, and R. S. Wallace, “Automatic Creation of 3D Facial Models”, IEEE Computer Graphics and Applications, 13(5), pp.16–22, 1993.

[2]M. Bichsel and A.P. Pentland, “A Simple Algorithm for Shape from Shading”, IEEE Proc. CVPR, pp. 459-465, 1992.

[3]V. Blanz and T. Vetter, “A Morphable Model for the Synthesis of 3D Faces”, SIGGRAPH, pp. 187–194, 1999.

[4]R.A. Dwyer, “A Faster Divide-and-Conquer Algorithm for Constructing Delaunay Triangulation”, Algorithmica, 2:137-151, 1987.

[5]P. Fua, “Regularized bundle-adjustment to model heads from image sequences without calibration data”, International Journal of Computer Vision, July 2000, pp. 153-171.

[6]B. Guentery, C. Grimmy, D. Woodz, H. Malvary, and F. Pighinz, “Making Faces”, SIGGRAPH, pp.55-66, 1998

[7]Reza Hassanpour, Volkan Atalay, “Delaunay Triangulation based 3D Human Face Modeling from Uncalibrated Images”, Computer Vision and Pattern Recognition Workshop, 2004 Conference on 27-02 June 2004 Page(s):75 – 75

[8]T. Horprasert, Y. Yacoob, and L.S. Davis, “Computing 3-D Head Orientation from a Monocular Image Sequence”, IEEE Proc. International Conference. AFGR, 1996.

[9]H. HS Ip and L.J. Yin, “Constructing 3D Individualized Head Model from Two Orthogonal Views”, The Visual Computer, 12(5), pp. 254-266, 1996.

[10]K.M. Lee and C.C.J. Kuo, “Shape from Shading with a Linear Triangular Element Surface Model”, IEEE Transactions on PAMI, 15(8), pp. 815-822, 1993.

[11]Y.C. Lee, D. Terzopoulos, and K. Waters, “Constructing Physics-Based Facial Models of Individuals”, In Proceedings of Graphics Interface, pp. 1–8, 1993.

[12]Z. Liu, Z. Zhang, C. Jacobs, and M. Cohen, “Rapid Modeling of Animated Faces From Video”, Microsoft Research, Technical Report, MSR-TR-2000-11.

[13]B.D. Lucas and T. Kanade, “An Iterative Image Registration Technique with an Application to Stereo Vision”, Proc. of DARPA IU, pp. 121-130, 1981.

[14]J. Maciel and J. Costeria, “Maximizing Rigidity: Optimal Matching under Scaled-Orthography”, ECCV, LNCS 2351, pp. 232-246, 2002.

[15]F. Pighin, J. Hecker, D. Lischinski, R. Szeliski, and D. H. Salesin, “Synthesizing Realistic Facial Expressions from Photographs”, SIGGRAPH, pp. 75–84, 1998.

[16]C. Poelman and T. Kanade, “A Paraperspective Factorization Method for Shape and Motion Recovery”, IEEE Transaction on PAMI, 19(3), pp. 206-218, 1997.

[17]A. Tankus, N. Sochen, and Y. Yeshurun, “A New Perspective [on] Shape-from-Shading”, IEEE ICCV, Vol. II, pp. 862–869, 2003.

[18]C. Tomasi and T. Kanade, “Shape and Motion from Image Streams under Orthography: a Factorization Method”, IJCV, 9(2), pp. 137-154, 1992.

[19]J.T. Twu and J.J. Lien, “Estimation of Facial Control-Point Locations” CVGIP E1-3, 2004.

[20]T.H. Wang and J.J. Lien, “Rigid and Non-Rigid Motion Separation Using 3D Model”, CVGIP, A2-5, 2004.

[21]R. Zhang, P.S. Tsai, J.E. Cryer, and M. Shah, “Shape from Shading: A Survey”, IEEE Transactions on PAMI, 21(8), pp.690–705, 1999.

[22]Q. Zheng and R. Chellappa, “Estimation of Illuminant Direction, Albedo, and Shape from Shading”, IEEE Transaction on PAMI, 13(7), pp. 680-702, 1991.