立體人臉模型可以提供給醫師做為病患手術顏面變化之預測用，本研究提出改良現有光影重建法(Shape-from-shading)之影像擷取系統，首先應用發光二極體陣列製作模擬一平面光源，以符合光影重建法之假設；同時使用850nm波長紅外光線，以降低環境中可見光源之干擾；其次運用同軸光拍攝方式，以袪除陰影所造成之影響；結合上述方法建構出了一符合光影重建法的符合實際應用之立體影像擷取系統。至於在重建演算法方面，則使用黏性解(Viscosity solution)方式以求解光影重建法問題，並使用快速步進法(Fast marching method)加速演算法之運算。根據本研究之實驗結果，應用本研究所提出方法與系統以重建之立體人臉假體外廓，與電腦斷層系統重建之人臉模型間的方均根誤差為5.6 mm，誤差分布在15.2至-12.5 mm之間；最後實際應用於真正人臉外廓，可知其可行性。

From the 3D facial model, clinician can easily use it as a tool to predict patient’s facial morphological change before surgery. This study proposes and presents an improved image acquisition system using shape-from-shading technique. Firstly, a rectangular arrangement of LED array is designed and constructed to simulate a flat type of light source to meet the assumption of shape-from-shading technique. Secondly, the 850 nm wavelength light within the IR spectrum is used to alleviate the interference problem of ambient light. Thirdly, the co-axes of radiant and reflected light concept is applied to eliminate shadows effect. Combining all the above-mentioned methods, an improved image acquisition system for 3D reconstruction of human face basing on the shape-from-shading technique is developed for real applications. In addition, the reconstruction algorithm integrating the viscosity solution approach is applied and implemented to solve shape from shading problem. Fast marching method is also used to speed up the computation of the algorithm. From the experimental results, the root mean square error between the proposed reconstruction method and CT reconstruction method for human head phantom is about 5.6 mm. The errors ranges from 15.2 to -12.5 mm. Finally, the proposed system is applied for real human face reconstruction for its feasibility.

[1] K. Cheung, S. Baker, and T. Kanade, “Shape-from-silhouette across time part i: Theory and algorithms,” International Journal of Computer Vision, vol. 62, pp. 221-247, 2005.
[2] A. Crouzil, X. Descombes, and J.D. Durou, “A Multiresolution Approach for Shape from Shading Coupling Deterministic and Stochastic Optimization,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 25, pp. 1416-1421, 2003.
[3] P. Daniel and J.D. Durou, “Creation of real images which are valid for the assumptions made in shape from shading,” 10th International Conference on Image Analysis and Processing, pp. 418-423, 1999.
[4] J.D. Durou, M. Falcone, and M. Sagona, “Numerical methods for shape-from-shading: A new survey with benchmarks,” Computer Vision and Image Understanding, vol. 109, pp. 22-43, 2008.
[5] M. Falcone and M. Sagona, “An algorithm for the global solution of the shape-from-shading model,” Lecture Notes in Computer Science, vol. 1310, pp. 596-603, 1997.
[6] A. Grün, F. Remondino, and L. Zhang, “Photogrammetric reconstruction of the Great Buddha of Bamiyan, Afghanistan,” The Photogrammetric Record, vol. 19, pp. 177-199, 2004.
[7] B.K.P. Horn. Obtaining shape from shading information. In P.H. Winston, editor, The Psychology of Computer Vision. McGraw-Hill, New York, 1975.
[8] J. Kain and D.N. Ostrov, “Numerical shape-from-shading for discontinuous photographic images,” International Journal of Computer Vision, vol. 44, pp. 163-173, 2001.
[9] R. Kozera, “Uniqueness in shape from shading revisited,” Journal of Mathematical Imaging and Vision, vol. 7, pp. 123-138, 1997.
[10] K.M. Lee and C.C.J. Kuo, “Shape from shading with a linear triangular element surface model,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 15, pp. 815-822, 1993.
[11] A. Lobay and D.A. Forsyth, “Shape from texture without boundaries,” International Journal of Computer Vision, vol. 67, pp. 71-91, 2006.
[12] A.P. Pentland, “Linear shape from shading,” International Journal of Computer Vision, vol. 4, pp. 153-162, 1990.
[13] A.P. Pentland, “Local shading analysis,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 6, pp. 170-187, 1984.
[14] E. Prados, F. Camilli, and O. Faugeras, “A unifying and rigorous shape from shading method adapted to realistic data and applications,” Journal of Mathematical Imaging and Vision, vol. 25, pp. 307-328, 2006.
[15] E. Prados and O. Faugeras, “A generic and provably convergent shape-from-shading method for orthographic and pinhole cameras,” International Journal of Computer Vision, vol. 65, pp. 97-125, 2005.
[16] E. Prados and S. Soatto, “Fast marching method for generic shape from shading,” Lecture Notes in Computer Science, vol. 3752, pp. 320-331, 2005.
[17] F. Remondino and S. El-Hakim, “Image-based 3D Modeling: A Review,” The Photogrammetric Record, vol. 21, pp. 269-291, 2006.
[18] A. Saxena, S.H. Chung, and A. Ng, “Learning depth from single monocular images,” Advances in Neural Information Processing Systems, vol. 18, p. 1161, 2006.
[19] J.A. Sethian, “A fast marching level set method for monotonically advancing fronts.,” Proceedings of the National Academy of Sciences of the United States of America, vol. 93, pp. 1591-1595, 1996.
[20] M.F. Tappen, W.T. Freeman, and E.H. Adelson, “Recovering intrinsic images from a single image,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 27, pp. 1459-1472, 2005.
[21] W.H. Ting, “Computer-aided system for predicting the facial changes in orthognathic surgery,” Master Thesis, Institute of Biomedical Engineering, National Cheng Kung University, 2000.
[22] P.S. Tsai and M. Shah, “Shape from shading using linear approximation,” Image and Vision Computing, vol. 12, pp. 487-498, 1994.
[23] F.A. Van den Heuvel, “A line-photogrammetric mathematical model for the reconstruction of polyhedral objects,” Proceedings of SPIE, vol. 3641, pp. 60-71, 1999.
[24] C.W. Wang, “3D reconstruction of dental cast from multiple 3D images,” Master Thesis, Institute of Biomedical Engineering, National Cheng Kung University, 2008.
[25] T.R. Wang, “Anatomy-based force model for predicting facial morphological change,” Master Thesis, Institute of Biomedical Engineering, National Cheng Kung University, 2008.
[26] Y. Weiss, “Deriving intrinsic images from image sequences,” Proceedings of Eighth IEEE International Conference on Computer Vision, vol. 2, pp. 68-75, 2001.
[27] R. Zhang, P.S. Tsai, J.E. Cryer, and M. Shah, “Shape-from-shading: a survey,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 21, pp. 690-706, 1999.