改進視角為基礎的AAM模型以用於建構人臉表情影像

簡易檢索 / 詳目顯示

回結果列表

研究生：	李柏蒼 Po-Tsang Li
論文名稱：	改進視角為基礎的AAM模型以用於建構人臉表情影像 Modified View-Based Active Appearance Model for Generation of Facial Image with Expression
指導教授：	黃仲陵 Chung-Lin Huang
口試委員:
學位類別：	碩士 Master
系所名稱：	電機資訊學院 - 電機工程學系 Department of Electrical Engineering
論文出版年：	2007
畢業學年度：	96
語文別：	英文
論文頁數：	36
中文關鍵詞：	人臉表情
外文關鍵詞：	AAM, facial expression
相關次數：	點閱：2 下載：0
分享至:	分享至facebook 分享至twitter

查詢本校圖書館目錄查詢臺灣博碩士論文知識加值系統勘誤回報

摘要
近年來，對於二維影像的人臉辨識技術已經達到令人滿意的結果。但有個廣為人知的問題，二維的人臉辨識率會因為人臉的角度變化而大幅衰減。我們雖然可以透過建立完整的3D 人臉模型來達到這個目的。但建立 3D 人臉模型的硬體需求高，又具有表情變化的模型需要儲存更大量的資料，處理時也需要更多的運算量，都是他不易實用的問題。如何找到有效率的替代方案，就們這篇文想探討的地方。
在此篇論文中，我們引用了 View-Based AAM 的概念，以數個不同角度範圍的 2D 人臉模型來涵蓋頭部的可能旋轉範圍。透過由事前資料所學得的，每個 2D 模型中人臉參數與角度的對應關係，和每個 2D 模型間參數的轉換關係，我們可以在找到輸入影像的人臉參數後，在另一個未見過的視角將他合成出來。為了不增加演算法的運算量，我們秉持 View-Based AAM 的精神以線性運算和PCA的方式來達到我們的目的。在這篇論文中我們延伸了表情空間在 View-Based AAM 中的應用，使其由原本應用於水平方向上各角度的的人臉，提昇為可具有各種表情變化的人臉影像上，並預測其他角度的圖像。
我們的實驗結果顯示，在使用Intel C2D 6300 處理器下，我們處理320*240大小的圖片，大約需要35~45 ms去預測其正面影像。

This thesis develops a method to solve the unpredictable head orientation problem in 2D facial analysis. In this thesis, we extend the expression subspace to the view-based AAM so that it can be applied for multi-view face fitting and pose correction for an input face of any expression. The experimental results will demonstrate that the proposed algorithm can be applied to improve the following facial identification process. We testing our system of the sequence by using Intel C2D 6300 CPU and the frame size is 320*240 pixels. It requires 30~45 ms to fitting a face and 0.35~0.45 ms for warping.

CONTENTS
CONTENTS……………………………………………………………II
LIST OF FIGURES…………………………………………………IV
LIST OF TABLES……………………………………………………IV
CHAPTER 1  INTRODUCTION……………………………..………1
1.1    Background  1
1.2    Related Works  2
1.3    Overview of the Thesis  3
1.4    Organization of the Thesis  5
CHAPTER 2  ACTIVE APPEARANCE MODEL…………………..6
2.1    Statistical Models of Appearance  6
2.2    Training Data  6
2.3    Shape Model  7
2.4    Texture Model  8
2.5    Appearance Model  10
2.6    Shape parameter weight  11
2.7    Active Appearance Model Search  11
CHAPTER 3  MODIFIED VIEW-BASED AAM…………………..14
3.1    Introduction  14
3.2    Multi-Pose 2D AAM Training  14
3.2.1    Training Data  14
3.3    Intra-Model Rotate 17
3.4    Identity and Expression Subspace  19
3.5    Inter-Model Rotate  21
3.6    Reconstruct a New View  22
CHAPTER 4  EXPERIMENTAL RESULTS AND DISCUSSIONS
              ………………………………………………………..24
4.1    Experimental：warping image vs Ground Truth  24
4.2    Performance Evaluation  28
4.3    Improvement of facial recognition  28
CHAPTER 5  CONCLUSIONS AND FUTURE WORKS………...32
REFERENCES………………………………………………………...33





LIST OF FIGURES
Fig. 1.1  The flowchart of our system..........…………………………………………5
Fig. 2.1  Examples of the training set…….………………………………………….7
Fig. 2.2  First two modes of shape variation(  sd)……….………………………8
Fig. 2.3  First two modes of texture variation(  sd)…….………………………..9
Fig. 2.4  First two modes of appearance variation(  sd)……..…………………11
Fig. 3.1  System environment………………………………………………………15
Fig. 3.2  Examples from the training sets for the models….……………………….16
Fig. 3.3  predicted angle vs actual angle across training set (a) result of our data. (b) Cootes’ experimental at ‘view-based active appearance mode’………………….18
Fig. 3.4  some examples from expression component training set..………………..19
Fig. 3.5  show the neutral image for training identity subspace….………………...20
Fig. 3.6  show the facial space relate to identity and expression...…………………21
Fig. 3.7  The flowchart of Rotate Model……..…………………………………...23
Fig. 4.1  result of warping Right Half to Frontal vs Ground truth.…………………26
Fig. 4.2  The experimental results of warping right-side view to frontal view….….27
Fig. 4.3  The experimental results of warping the right-half-view facial image to front view………………………………………………………………………...…..29
Fig. 4.4 The experimental results of warping the right-side-view facial image to the front view…………………………………………………………………………...30
LIST OF TABLES
Table 5.1  Process time of each steps in our system.………….……………………28
Table 5.2  The improvement of identity recognition, with ICO (identity component only) and PC (pose correction) with 15 degree….…………….……………………31

                                

REFERENCES
[1] T.F. Cootes, D. Cooper, C.J. Taylor and J. Graham, "Active Shape Models - Their Training and Application." Computer Vision and Image Understanding. Vol. 61, No. 1, Jan. 1995, pp. 38-59.
[2] T.F.Cootes, G.J. Edwards and C.J.Taylor. "Active Appearance Models", in Proc. European Conference on Computer Vision 1998 (H.Burkhardt & B. Neumann Ed.s). Vol. 2, pp. 484-498, Springer, 1998.
[3] G.J.Edwards, C.J.Taylor, T.F.Cootes, "Interpreting Face Images using Active Appearance Models", Int. Conf. on Face and Gesture Recognition 1998. pp. 300-305
[4] T. F. Cootes, G.V.Wheeler, K.N.Walker and C. J. Taylor, "View-Based Active Appearance Models", Image and Vision Computing, Vol.20, 2002, pp.657-664
[5] T.F. Cootes, G.V.Wheeler, K.N. Walker and C.J.Taylor "Coupled-View Active Appearance Models", Proc. British Machine Vision Conference 2000, vol.1,pp.52-61
[6] T.F.Cootes, G.J. Edwards and C.J.Taylor. "Active Appearance Models", IEEE PAMI, Vol.23, No.6, pp.681-685, 2001
[7] H. Kang, T.F. Cootes and C.J. Taylor, "A Comparison of Face Verification Algorithms using Appearance Models", Proc.BMVC2002, Vol.2,pp.477-4862.

[8] M. B. Stegmann, B. K. Ersbøll, R. Larsen, "FAME -- A Flexible Appearance Modelling Environment", IEEE Transactions on Medical Imaging, IEEE, 2003
[9] I. Matthews and S. Baker. “Active Appearance Models revisited.” IJCV, 2004. In Press
[10] 陳曉瑩 "即時多角度人臉偵測" 國立清華大學電機工程研究所碩士論文,2006
[11] V. Blanz and T. Vetter. "A morphable model for the synthesis of 3d faces. " Con. Proc. Computer Graphics SIGGRAPH '99, 1999.
[12] V. Blanz and T. Vetter. "Face recognition based on fitting a 3d morphable model. " IEEE trans. On Pattern Analysis and Machine Intelligence, 25(9), September 2003.
[13] C. Christoudias, L. Morency, and T. DarreIl. "Light field appearance manifolds. " Proc. of the European Conference on Vision, (4):482-493, 2804. Computer Vision, (4):481-493, 2004
[14] R. Gross, I. Matthews, and S. Baker. "Eigen light-fields and face recognition across pose. " International Conf On Automatic Face and Gesture Recognition, 2002.

[15] Chang, Jianglong, Zheng, Ying, Wang, Zengfu "Facial Expression Anaylsis and synthesis : a Bilinear Appraoach" Information Acquisition, 2007, ICIA ’07 International Conference on 8-11 July 2007 Page(s):457 – 464
[16] Wang, Tueming, Pen, Gang, Wu, Zhaohui "3D Face Recognition in the Presence of Expression : A Guidance-based Constraint Deformation Approach" Computer Vision and Pattern Recognition, 2007, CVPR ’07, IEEE Conference on 17-22 June 2007 Page(s):1-7
[17] Amor, B.B. Ardabilianm, M., Chen, L. "New Experiments on ICP-Based 3D Face Recognition and Authentication" Pattern Recognition, 2006, ICPR 2006. 18th international Conference on Volume 3, 0-0 0 Page(s) : 1195 – 1199
[18] Kakadiaris I.A., Passalis, G., Toderici, G., Murtuza, M.N., Yunliang Lu, Karampatziakis, N, Theoharis, T. "Three-Dimensional Face Recognition in the Presence of Facial Expression: An Annotated Deformable Model Approach" Pattern Analysis and Machine Intelligence. IEEE Transactions on Volume 29, Issue 4, April 2007 Page(s) : 640-649
[19] Ramanathan, S., Kassim, A., Vemlatesh, Y.V., Wu Sin Wah, "Human Facial Expression Recognition using a 3D Morphable Model" Image Processing, 2006 IEEE international Conference on 8-11 Oct 2006 Page(s):661-664
[20] Lu Xiaoguang, Jain Anil, "Deformation Modeling for Robust 3D face Matching" IEEE Transactions on Pattern Analysis and Machine Intelligence : Accepted for future publication Volume PP, Issue 99, 2007 Page(s): 1-1
[21] Jling Xiao, Baker, S., Mathews, I., Kanade, T. "Real-time conbined 2D+3D active appearance models" Computer Vision and Pattern Recognition, 2004 CVPR 2004. Proceedings of the 2004 IEEE Computer Society Conference on Volume 2, 27 June-2 July 2004 Page(s):II-535 – II-542 Vol.2
[22] Koterba, S., Baker, S., Mathews, I., Changbo Hu, Jing Xiao, Cohn, J., Janade, T. "Multi-view AAM fitting and camera calibration" Computer Vision, 2005. ICCV 2005. Tenth IEEE International conference on Volume 1, 17-21 Oct. 2005 Page(s):511-518 Vol.1
[23] Jaewon Sung, Daijin Kim "STAAM : Fitting a 2D+3D to Stereo Images" Image Processing, 2006 IEEE international Conference on 8-11 Oct. 2006 Page(s): 2781 – 2784
[24] Lucey, S., Mathews, I., Changbo Hu, Ambadar, Z., de la Torre, F., Cohn, J., "AAM derived face representations for robust action recognition" Automatic Face and Gesture Recognition. 2006. FGR 2006. 7th internation Conference on 10-12 April 2006 Page(s): 155-160
[25] Huisman, P., van Munster, R., Moro-Ellenberger, S., Veldhuis, R., Bazen, A. "Making 2D face recognition more robust using AAMs for pose compensation" Automatic Face and Gesture Recognition, 2006. FGR 2006. 7th international Conference on 10-12 April 2006 Page(s):6 pp.

全文公開日期本全文未授權公開 (校內網路)
全文公開日期本全文未授權公開 (校外網路)

簡易檢索 / 詳目顯示

相關論文