隨著智慧生活的發展，識別技術是當中十分重要的一環，尤其以人臉作為生物特徵識別的主要趨勢。人臉辨識技術不僅在學術界和產業界都是一個炙手可熱的議題。然而，在未經限制的環境下，如光線變化、頭部轉動等情形，辨識準確率容易受其影響。其中，頭部的轉動會嚴重的降低辨識率，也是最無法控制的環境因素。本篇論文中，我們架構一套完整人臉辨識系統，包括人臉偵測、人臉校正以及辨識技術，並以人臉校正的改善作為本論文主軸，以解決上述提及之問題。因此，將針對舊有的精確形狀回歸法進行改善，提出結合特徵形狀回歸法，利用像素差異性、區域像素模組差異性、及梯度大小差異性於人臉上選取的七個特定標的，包含眼睛、鼻子、嘴巴，使其廣義的人臉輪廓描述更為符合實際情形。然後利用提出的條件式十字變形法將測試人臉校正為正臉，以避免了部分的校正誤差。最後，在實驗上分為兩大主軸，其一是與舊系統校正錯誤率之比較結果，並證明提出的結合特徵形狀回歸法與條件式十字變形法能更符合實際廣義人臉的定義；其二是證明在具頭部轉動的情況下，我們提出的方法能對最後辨識結果有明顯的助益。

With the growth of the smart living, the intelligent identification becomes more and more important. Especially, the facial feature is a most popular biometric for person recognition. Face recognition develops all the range not only in academic community but also in industrial circles. Nevertheless, the accuracy would be degraded seriously when the algorithms operated in unconstraint environments such as lighting changes, pose variations. One of the most common issue is the pose variation, since the head rotation influences the recognition performance a lot yet it is evitable. In this thesis, a face recognition system is proposed including face detection, face alignment and face recognition modules. The face alignment is a major issue in this thesis for solving the mentioned problems. Thus, the joint feature shape regression (JFSR) is proposed to improve the defects in conventional explicit shape regression (ESR). After the selected 7 landmark positions, including canthus, nose tip, and the corner of the mouth are acquired, the conditional cross-based warping method is proposed to affine the face and refrain some alignment errors simultaneously. Experimental results are divided into two parts. First, the JFSR are compared with the ESR for general face shape regression. The result shows that the proposed JFSR is more suitable than the ESR for retrieving global facial shape. Secondly, our proposed system is revealed that the recognition accuracy can be improved under head rotation variations.

[1] Yang, G. and T.S. Huang, Human face detection in a complex background. Pattern Recognition, 1994. 27(1): p. 53-63.
[2] Hotta, K., T. Kurita, and T. Mishima. Scale invariant face detection method using higher-order local autocorrelation features extracted from log-polar image. in Automatic Face and Gesture Recognition, 1998. Proceedings. Third IEEE International Conference on. 1998. IEEE.
[3] Han, C.-C., et al., Fast face detection via morphology-based pre-processing. Pattern Recognition, 2000. 33(10): p. 1701-1712.
[4] Fasel, B., Fast multi-scale face detection. 1998, IDIAP.
[5] Rowley, H.A., S. Baluja, and T. Kanade, Neural network-based face detection. IEEE Transactions on pattern analysis and machine intelligence, 1998. 20(1): p. 23-38.
[6] Viola, P. and M.J. Jones, Robust real-time face detection. International journal of computer vision, 2004. 57(2): p. 137-154.
[7] Lienhart, R. and J. Maydt. An extended set of haar-like features for rapid object detection. in Image Processing. 2002. Proceedings. 2002 International Conference on. 2002. IEEE.
[8] Kotropoulos, C. and I. Pitas. Rule-based face detection in frontal views. in Acoustics, Speech, and Signal Processing, 1997. ICASSP-97., 1997 IEEE International Conference on. 1997. IEEE.
[9] Birchfield, S. An elliptical head tracker. in Signals, Systems & Computers, 1997. Conference Record of the Thirty-First Asilomar Conference on. 1997. IEEE.
[10] Wu, H., Q. Chen, and M. Yachida, Face detection from color images using a fuzzy pattern matching method. IEEE transactions on pattern analysis and machine intelligence, 1999. 21(6): p. 557-563.
[11] Cootes, T.F. and C.J. Taylor, Active shape models—‘smart snakes’, in BMVC92. 1992, Springer. p. 266-275.
[12] Cristinacce, D. and T.F. Cootes. Boosted regression active shape models. in BMVC. 2007.
[13] Cootes, T.F., G.J. Edwards, and C.J. Taylor. Active appearance models. in European conference on computer vision. 1998. Springer.
[14] Matthews, I. and S. Baker, Active appearance models revisited. International Journal of Computer Vision, 2004. 60(2): p. 135-164.
[15] Sauer, P., T.F. Cootes, and C.J. Taylor. Accurate Regression Procedures for Active Appearance Models. in BMVC. 2011.
[16] Saragih, J. and R. Goecke. A nonlinear discriminative approach to AAM fitting. in 2007 IEEE 11th International Conference on Computer Vision. 2007. IEEE.
[17] Dollár, P., P. Welinder, and P. Perona. Cascaded pose regression. in Computer Vision and Pattern Recognition (CVPR), 2010 IEEE Conference on. 2010. IEEE.
[18] Valstar, M., et al. Facial point detection using boosted regression and graph models. in Computer Vision and Pattern Recognition (CVPR), 2010 IEEE Conference on. 2010. IEEE.
[19] Cao, X., et al., Face alignment by explicit shape regression. International Journal of Computer Vision, 2014. 107(2): p. 177-190.
[20] Turk, M. and A. Pentland, Eigenfaces for recognition. Journal of cognitive neuroscience, 1991. 3(1): p. 71-86.
[21] Belhumeur, P.N., J.P. Hespanha, and D.J. Kriegman, Eigenfaces vs. fisherfaces: Recognition using class specific linear projection. IEEE Transactions on pattern analysis and machine intelligence, 1997. 19(7): p. 711-720.
[22] Moghaddam, B. and A. Pentland, Probabilistic visual learning for object representation. IEEE Transactions on pattern analysis and machine intelligence, 1997. 19(7): p. 696-710.
[23] Yang, J., et al., Two-dimensional PCA: a new approach to appearance-based face representation and recognition. IEEE transactions on pattern analysis and machine intelligence, 2004. 26(1): p. 131-137.
[24] Shawe-Taylor, J. and N. Cristianini, Kernel methods for pattern analysis. 2004: Cambridge university press.
[25] Martínez, A.M. and A.C. Kak, Pca versus lda. IEEE transactions on pattern analysis and machine intelligence, 2001. 23(2): p. 228-233.
[26] Schölkopf, B., A. Smola, and K.-R. Müller, Nonlinear component analysis as a kernel eigenvalue problem. Neural computation, 1998. 10(5): p. 1299-1319.
[27] Yang, M.-H. Kernel Eigenfaces vs. Kernel Fisherfaces: Face Recognition Using Kernel Methods. in fgr. 2002.
[28] Scholkopft, B. and K.-R. Mullert, Fisher discriminant analysis with kernels. Neural networks for signal processing IX, 1999. 1(1): p. 1.
[29] Baudat, G. and F. Anouar, Generalized discriminant analysis using a kernel approach. Neural computation, 2000. 12(10): p. 2385-2404.
[30] Lu, J., K.N. Plataniotis, and A.N. Venetsanopoulos, Face recognition using kernel direct discriminant analysis algorithms. IEEE Transactions on Neural Networks, 2003. 14(1): p. 117-126.
[31] Huang, J., et al., Choosing parameters of kernel subspace LDA for recognition of face images under pose and illumination variations. IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics), 2007. 37(4): p. 847-862.
[32] Naseem, I., R. Togneri, and M. Bennamoun, Linear regression for face recognition. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2010. 32(11): p. 2106-2112.
[33] Liu, G., Y. Yan, and H. Wang. Robust Modular Linear Regression Based Classification for Face Recognition with Occlusion. in Image and Graphics (ICIG), 2013 Seventh International Conference on. 2013. IEEE.
[34] Freund, Y., R. Schapire, and N. Abe, A short introduction to boosting. Journal-Japanese Society For Artificial Intelligence, 1999. 14(771-780): p. 1612.
[35] Duffy, N. and D. Helmbold, Boosting methods for regression. Machine Learning, 2002. 47(2-3): p. 153-200.
[36] Friedman, J.H., Greedy function approximation: a gradient boosting machine. Annals of statistics, 2001: p. 1189-1232.
[37] Ozuysal, M., et al., Fast keypoint recognition using random ferns. IEEE transactions on pattern analysis and machine intelligence, 2010. 32(3): p. 448-461.
[38] Soriano, M., et al. Using the skin locus to cope with changing illumination conditions in color-based face tracking. in IEEE Nordic Signal Processing Symposium. 2000. Kolmarden, Sweden: IEEE.
[39] Belhumeur, P.N., et al., Localizing parts of faces using a consensus of exemplars. IEEE transactions on pattern analysis and machine intelligence, 2013. 35(12): p. 2930-2940.
[40] Le, V., et al. Interactive facial feature localization. in European Conference on Computer Vision. 2012. Springer.
[41] Martinez, A.M., The AR face database. CVC Technical Report, 1998. 24.
[42] Phillips, P.J., et al. Overview of the face recognition grand challenge. in 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05). 2005. IEEE.
[43] Kazemi, V. and J. Sullivan. One millisecond face alignment with an ensemble of regression trees. in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2014.
[44] Burgos-Artizzu, X.P., P. Perona, and P. Dollár. Robust face landmark estimation under occlusion. in Proceedings of the IEEE International Conference on Computer Vision. 2013.
[45] Xiong, X. and F. De la Torre. Supervised descent method and its applications to face alignment. in Proceedings of the IEEE conference on computer vision and pattern recognition. 2013.
[46] Wright, J., et al., Robust face recognition via sparse representation. IEEE transactions on pattern analysis and machine intelligence, 2009. 31(2): p. 210-227.
[47] Jiang, X. and J. Lai, Sparse and dense hybrid representation via dictionary decomposition for face recognition. IEEE transactions on pattern analysis and machine intelligence, 2015. 37(5): p. 1067-1079.
[48] He, X., et al., Face recognition using Laplacianfaces. IEEE transactions on pattern analysis and machine intelligence, 2005. 27(3): p. 328-340.
[49] He, X., et al. Neighborhood preserving embedding. in Tenth IEEE International Conference on Computer Vision (ICCV'05) Volume 1. 2005. IEEE.
[50] Huang, S.-M. and J.-F. Yang, Improved principal component regression for face recognition under illumination variations. IEEE signal processing letters, 2012. 19(4): p. 179-182.
[51] Huang, S.-M. and J.-F. Yang, Unitary regression classification with total minimum projection error for face recognition. IEEE signal processing letters, 2013. 20(5): p. 443-446.
[52] Huang, S.-M. and J.-F. Yang, Linear discriminant regression classification for face recognition. IEEE Signal Processing Letters, 2013. 20(1): p. 91-94.
[53] Chou, Y.-T., S.-M. Huang, and J.-F. Yang, Class-specific kernel linear regression classification for face recognition under low-resolution and illumination variation conditions. EURASIP Journal on Advances in Signal Processing, 2016. 2016(1): p. 1-9.