臺灣博碩士論文加值系統

　　這篇論文中提出一個以外貌為基底的多角度人臉偵測及角度估計的系統。它是一種由簡單到複雜的架構，配合全體到區域性的特徵使用，並且達到粗略及精細的角度估計。首先，串接式的非人臉移除器在特徵空間中採用全體的資訊達到移除非人臉的測試樣本。在串接移除器的最後一個階段，所包含的人臉模型有隱含人臉角度的資訊，借此做為粗略的角度估計。此粗估的角度像是一個多功器的選擇訊號，將通過串接移除器的測試樣本送至下一個功能狀態－針對不同角度的貝式分類器。此貝式分類器採用區域性的特徵，並且以主成分分析的成分及獨立分析的成分為基底，統計人臉及非人臉樣本的模型。透過統計後建立的模型，分類器將人臉樣本和非人臉樣本做分辨。最後，根據統計後的角度模型，進行精細的角度估計。實驗顯示每個功能狀態的成果，以及整個系統所達到的低錯誤率及高正確率的結果。

　We present an appearance-based technique for not only detecting multi-view faces but also estimating the corresponding poses. Our architecture concerns the global-to-local facial features and estimates the coarse-to-fine pose angle. The system is composed of four subsystems. Firstly, for the cascaded nonface rejecter subsystem, the goal of the nonface rejecters is to reject most nonface patterns (20x20-pixel window) in order to reduce the computational time of remaining two subsystems. Secondly, based on the projection weights of the global facial feature eigenspace, the coarse pose angle of each survived face or nonface patterns is estimated. Thirdly, for the view-based face detector subsystem, the view-based detectors use the joint probability of local features and corresponding positions to model the face. Each local feature projects into the corresponding eigenspace and the ‘residual independent basis’ space. Then Bayes decision rule is applied to judge whether the input pattern contains a face or not. Finally, for the fine pose estimate subsystem, the fine pose angle of the detected face is estimated by using the combination of vector quantization and maximum likelihood probability. According to the experimental results, the accuracy of the face detection and the result of pose estimation are promising.

CHAPTER 1. INTRODUCTION 1
CHAPTER 2. RELATED WORKS 3
CHAPTER 3. SYSTEM OVERVIEW 6
3.1 Cascaded Nonface Rejecters 8
3.2 Coarse Pose Estimation 10
3.3 View-Based Detectors 12
3.4 Fine Pose Estimation 18
CHAPTER 4. TESTING PROCESS 20
CHAPTER 5. EXPERIMENTAL RESULTS 21
5.1 Performance of the Cascaded Nonface Rejecters 21
5.2 Performance of the View-based Detectors 23
5.3 Accuracy of Pose Estimation Subsystem 26
5.4 Testing Performance on the Entire System 27
CHAPTER 6. CONCLUSION 34
REFERENCE 35

[1] M.S. Barlett, J.R. Movellan, and T.J. Sejnowski, “Face Recognition by ICA”, IEEE Trans. on Neural Network, Vol. 13, pp 1450-1464, 2002.
[2] H.B. Barlow, “Unsupervised learning”, Neural Computation, Vol.1, pp. 295-311, 1989.
[3] S.M. Bileschi and B. Heisele, “Advances in Component Based Face Detection”, In Proc. IEEE Workshop on Analysis and Modeling on Faces and Gestures, pp. 149-156, 2003.
[4] M.C. Bural, U.M. Fayyad, P. Perona, P. Smyth, and M.P. Burl, “Automating the Hunt for Volcanoes on Venus”, In Proc. IEEE Conf. on CVPR, pp. 302-309, 1994.
[5] J. Feraud, Or. Bernier, and M. Collobert, “A Fast and Accurate Face Detector for Indexation of Face Images”, In IEEE Int’l. Conf. on FG, pp. 77-82, 2000.
[6] C. Garcia and M. Delakis, “A Neural Architecture for Fast and Robust Face Detection”, In Proc. IEEE Int’l. Conf. on Pattern Recognition (PR), Vol. 2, pp. 44-47, 2002.
[7] C. Garcia and M. Delakis, “Convolutional Face Finder: A Neural Architecture for Fast and Robust Face Detection”, IEEE PAMI, Vol. 1, pp.1408-1423,2004.
[8] S. Gong, S. McKenna, and J. Collins, “An Investigation into Face Pose Distribution”, In IEEE Int’l. Conf. on Automatic Face and Gesture Recognition (FG), pp. 265-270, 1996.
[9] B. Heisele, P. Ho, J. Wu, and T. Poggio, “Face Recognition: Component-based versus Global Approaches”, IEEE Conf. on CVIU, Vol. 91, pp. 302-309, 1994.
[10] B. Heisele, T. Serre, M. Pontil and T. Poggio, “Component-based Face Detection”, In Proc. IEEE Conf. on CVPR, Vol. 1, pp. 657-662, 2001.
[11] A. Hyvarinen and E. Oja, “Independent Component Analysis: A Tutorial”, 1999.
[12] J. Huang, X. Shao, and H. Wechsler, “Face Pose Discrimination Using Support Vector Machine”, IEEE Int’l. Conf. on PR, Vol.1, pp. 154-156, 1998.
[13] T. Kato, Y. Ninomiya, and I. Masaki, ”Preceding Vehicle Recognition Based on Learning from Sample Images”, IEEE Trans. on Intelligent Transportation System, Vol. 3, pp. 252-260, 2002.
[14] T.K. Kim, H. Kim, W. Hwang, and S.C. Kee, “Independent Component Analysis in a Facial Local Residue Space”, In Proc. IEEE Conf. on CVPR, pp. 579-786 2003.
[15] T.W. Lee, M. Girolami and T.J. Sejnowski, “Independent Component Analysis using an Extended Informax Algorithm for Mixed Sub-Gaussian and Super-Gaussian Sources”, Neural Computation, Vol. 11, pp.417-441, 1999.
[16] K. Levi and Y. Weiss, “Learning Object Detection from a Small Number of Examples: The Importance of Good Features”, In Proc. IEEE Conf. on Computer Vision and Pattern Recognition (CVPR), Vol. 2, pp. 53-60, 2004.
[17] T.K. Leung, M.C. Burl, and P. Perona, “Finding faces in cluttered scenes using random labeled graph matching”, In Proc. IEEE Int’l. Conf. on ICCV, pp. 637-644, 1995.
[18] B. Leung, “Component-based Car Detection in Street Scene Images”, Master Thesis, Department of Electrical Engineering and Computer Science, MIT, May, 2004.
[19] S.Z. Li et al, “Learning Low Dimensional Invariant Signature of 3-D Object under Varying View and Illumination from 2-D Appearances”, In Proc. of Int’l. Conf. on Computer Vision, Vol. 1, pp. 635-640, 2001.
[20] S.Z. Li et al, “Kernel Machine Based Learning For Multi-View Face Detection and Pose Estimation”, In Proc. ICCV, pp.674-679, 2001.
[21] S.Z. Li et al, “Multi-View Face Pose Estimation Based on Supervised ISA Learning”, In Proc. IEEE Int’l. Conf. on FG, pp. 100-105, 2002.
[22] S.Z. Li et al, “FloatBoost Learning and Statistical Face Detection”, IEEE Trans. on Pattern Analysis an Machine Intelligence (PAMI), Vol. 26, pp. 1112-1123, 2004.
[23] C. Liu, “A Bayesian Discriminating Features Method for Face Detection”, IEEE PAMI Vol. 25, pp. 725-740, 2003.
[24] T. Mikami and M. Wada, “Example-based Face Detection using Independent Component Analysis and RBF Network”, SICE Conference, 2003.
[25] B. Moghaddam and A. Pentland, “Probabilistic Visual Learning for Object Representation”, IEEE PAMI, Vol.19, pp. 696-710, 1997.
[26] H. Murase and S.K. Nayar, “Visual Learning and Recognition of 3-D Objects from Appearance”, International Journal of Computer Vision, 14:5-24, 1995.
[27] J. Ng and S. Gong, “Performing Multi-View Face Detection and Pose Estimation Using a Composite Support Vector Machine across The View Sphere”, In Proc. of IEEE Int. Workshop on Recognition, Analysis, and Tracking of Faces and Gestures in Real-Time Systems, pp. 14–21, 1999.
[28] C.P. Papageorgiou, M. Oren and T. Poggio, “A general framework for object detection”, In Proc. ICCV, Vol. 1, pp. 555-562, 1998
[29] A. Pentland, B. Moghaddam, and T. Starner, “View-Based and Modular Eigenspaces of Face Recognition”, In Proc. IEEE Conf. on CVPR, pp. 84–91, 1994.
[30] E. Petajan, H.P. Graf, T. Chen, and E. Cosatto, “Locating Face and Facial Parts”, In Proc. IEEE FG, pp. 41-46, 1995.
[31] A.N. Rajagopalan, P. Burlina, and R. Chellapa, “High Order Statistical Learning for Vehicle Detection in Images”, In Proc. ICCV, Vo. 2, pp. 1204-1209, 1999
[32] H.A. Rowley, S. Baluja, and T. Kanade, “Neural Network-Based Face Detection”, IEEE PAMI, Vol. 20, pp. 22–38, 1998.
[33] H. Schneiderman and T. Kanade, “A Statistical Method for 3D Object Detection Applied to Faces and Cars”, In Proc. IEEE Conf. on CVPR, pp. 746–751, 2000.
[34] H. Schneiderman, “Feature-Centric Evaluation for Efficient Cascaded Object Detection”, In Proc. IEEE Conf. on CVPR, Vol.2, pp. 29-36, 2004.
[35] S. Srinivasan and K.L. Boyer, “Head Pose Estimation Using View Based Eigenspaces”, In Proc. IEEE Int’l. Conf. on PR, Vol. 4, pp. 302-305, 2002. S. E. Levinson, “Continuously variable duration hidden Markov models for automatic speech recognition”, Computer Speech and Language, vol. 1, pp. 29-45, 1986.
[36] Z. Sun, G Bebis, and R. Miller, “Object Detection Using Feature Subset Selection, “IEEE Int’l. Conf. on PR, Vol. 37, pp. 2165-2176, 2004.
[37] K.K. Sung and T. Poggio, “Example-Based Learning for View-Based Human Face Detection”, IEEE PAMI, pp. 39-51, 1998.
[38] P. Viola and M. Jones, “Rapid Object Detection Using a Boosted Cascade of Simple Features”, In Proc. IEEE Conf. on CVPR, pp. 511-518, 2001.
[39] P. Viola and M. Jones, “Fast Multi-View Face Detection”, Technical Report TR2003-96, Mitsubishi Electric Research Laboratories, 2003.
[40] C.A. Waring and X. Liu, “Face Detection Using Spectral Histograms and SVMs”, IEEE Trans. On Systems, Man, and Cybernetics, Vol. 35, 2005
[41] Y. Wei, L. Fradet, and T. Tan, “Head Pose Estimation Using Gabor Eigenspace Modeling”, In Proc. IEEE Int’l. Conf. on Image Processing, Vol. 1, pp. 281-284, 2002.
[42] L. Wiskott, J. Fellous, N. Kruger, and C.V. Malsburg, “Face Recognition By Elastic Bunch Graph Matching”, IEEE PAMI, Vol. 19, pp. 775-779, 1997.
[43] B. Wu, H. Ai, C. Huang, and S. Lao, “Fast Rotation Invariant Multi-view Face Detection Based on Real Adaboost”, In Proc.IEEE Int’l. Conf. FG, pp. 79-84, 2004.
[44] R. Xiao, M.J. Li and H.J. Zhang, “Robust Multipose Face Detection in Images”, IEEE Trans. on Circuits and System for Video Technology, Vol. 14, pp. 31-41, 2004.
[45] M.H. Yang, D.J. Kriegman, and N. Ahuja, “Detecting Faces in Images: A Survey”, IEEE PAMI, Vol. 24, pp.34-58, 2002.
[46] Z. Yang, H. Ai, B. Wu, S. Lao, and L. Cai, “Face Pose Estimation and Its Application in Video Shot Selection”, In Proc. IEEE Int’l. Conf. on PR, pp. 322-325, 2004.