特徵轉換結合度量尺度學習與其在步態辨識上之應用

以作者查詢圖書館館藏

、以作者查詢臺灣博碩士

、以作者查詢全國書目

、勘誤回報

、線上人數：27

、訪客IP：18.227.13.169

姓名

李俊傑(Chun-chieh Lee) 查詢紙本館藏

畢業系所

資訊工程學系

論文名稱

特徵轉換結合度量尺度學習與其在步態辨識上之應用
(Feature Transformation Coupled with Metric Learning with Application to Gait Recognition)

相關論文

★ 使用視位與語音生物特徵作即時線上身分辨識	★ 以影像為基礎之SMD包裝料帶對位系統
★ 手持式行動裝置內容偽變造偵測暨刪除內容資料復原的研究	★ 基於SIFT演算法進行車牌認證
★ 基於動態線性決策函數之區域圖樣特徵於人臉辨識應用	★ 基於GPU的SAR資料庫模擬器：SAR回波訊號與影像資料庫平行化架構 (PASSED)
★ 利用掌紋作個人身份之確認	★ 利用色彩統計與鏡頭運鏡方式作視訊索引
★ 利用欄位群聚特徵和四個方向相鄰樹作表格文件分類	★ 筆劃特徵用於離線中文字的辨認
★ 利用可調式區塊比對並結合多圖像資訊之影像運動向量估測	★ 彩色影像分析及其應用於色彩量化影像搜尋及人臉偵測
★ 中英文名片商標的擷取及辨識	★ 利用虛筆資訊特徵作中文簽名確認
★ 基於三角幾何學及顏色特徵作人臉偵測、人臉角度分類與人臉辨識	★ 一個以膚色為基礎之互補人臉偵測策略

檔案

[Endnote RIS 格式]

[Bibtex 格式]

[相關文章]

[文章引用]

[完整記錄]

[館藏目錄]

[檢視]

[下載]

本電子論文使用權限為同意立即開放。
已達開放權限電子全文僅授權使用者為學術研究之目的，進行個人非營利性質之檢索、閱讀、列印。
請遵守中華民國著作權法之相關規定，切勿任意重製、散佈、改作、轉貼、播送，以免觸法。

摘要(中)

幾十年來，利用各種影像或視訊來進行於身分辨識的應用技術，有相當多的研究單位關注與投入。其中，行人步態被認為是一種很有潛力的特徵，它能在較遠距離或低解析度的視訊中來有效識別身分。在這篇論文中，我們提出了行人身分辨識的架構，此架構同樣是擷取行人步行時所呈現的特徵，但它能進一步處理當測試者與資料庫裡的人被拍攝的角度可能是不一樣的情形。在我們的架構中，原始步態特徵是一個從行走視訊所提取的時空模板向量。這些特徵向量先投影到相對應的子空間，此子空間與拍攝角度有關。然後，這些位於相同子空間的特徵向量，被用來學習屬於該子空間裡特有的度量尺度。
在測試者與資料庫的拍攝角度是相同的情形下，用事先在該子空間所學習得來的度量尺度來計算相似程度。另外，在測試者與資料庫的拍攝角度是不一樣的情形下，我們事先建構好視角轉換模型(VTM)。在辨識測試者的身分時，假設資料庫的拍攝角度為 j，測試者的拍攝角度為 i，先將測試者的子空間特徵向量，轉換投影到資料庫的拍攝角度 j 所對應的子空間裡。然後，測試者與資料庫裡行人間的相似程度，是用資料庫對應子空間所學習得來的度量尺度來計算。我們用公開的標竿步態資料庫進行多個實驗，實驗結果顯示，結合特徵轉換與度量尺度學習的技術，對行人身分識別率的提升有顯著的效果。

摘要(英)

Human identification using various visual cues has gained many research attentions for decades. Among these, gait feature has been considered as a promising way to recognize individual at a distance or at low resolution. In this dissertation, we propose the human recognition framework based on the biometric trait conveyed by a walking subject, where the viewing angles of gallery and probe may differ. The initial gait feature used in our framework is the spatial temporal template extracted from one walking sequence. These feature vectors are projected into the corresponding subspace respective to the capturing angle for the walking subject. Then, the embedding feature vectors, which are viewing angle dependent, are used to learn a distance metric.
In identical view gait recognition, the metric learned from the embedded vectors of the same view is employed for similarity measurement between the probe and gallery. In cross view gait recognition, where the viewing angles of the probe and gallery are different, a view transformation model (VTM) is constructed by learning scheme in advance. At the recognition stage, assuming the gallery set is collected at viewing angle j, the embedded vector of the probe captured at another viewing angle i will be firstly transformed into the subspace spanned by gallery embedded vectors. Then, the similarities between the probe and gallery are measured using the metric learned on the subspace corresponding to viewing angle j. Experiments were conducted on public benchmark database and the results demonstrate that notable improvement of gait recognition performance via the combination of feature transformation and metric learning is accomplished as anticipated.

關鍵字(中)

★ 步態辨識
★ 視角轉換模型
★ 度量尺度學習

關鍵字(英)

★ Gait Recognition
★ View Transformation Model
★ Metric Learning

論文目次

摘要 I
Abstract II
誌謝 III
List of Figures V
List of Tables IX
Chapter 1 Introduction 1
1.1 Motivation 1
1.2 Challenges 5
1.2.1 Feature Representation 6
1.2.2 Large Intra-class Variance 7
1.2.3 Feature Classification 8
1.3 Literature Review 10
1.4 Organization of Thesis 17
Chapter 2 Metric Learning and Identical View Gait Recognition 18
2.1 Learning Low-dimensional Manifold Subspace 20
2.2 Metric Learning on Manifold Subspace 26
2.3 The Framework for Identical View Gait Recognition 31
2.3.1 Template Generation Module 32
2.3.2 Manifold Learning Module 34
2.3.3 Metric Learning Module 35
2.3.4 Manifold Projection Module 35
2.3.5 Recognition Module 36
Chapter 3 View Transformation and Cross View Gait Recognition 37
3.1 The View Transformation Model 37
3.2 The Framework for Cross View Gait Recognition 40
3.2.1 The VTM Construction 41
3.2.2 Cross View Gait Recognition 42
3.2.3 Viewing Angle Determination 44
Chapter 4 Experimental Results 46
4.1 Experimental Results of Identical View Gait Recognition 48
4.2 Experimental Results of Cross View Gait Recognition 51
Chapter 5 Conclusions and Future Works 71
References 73

參考文獻

[1] D. Cunado, M.S. Nixon, J.N. Carter, "Using gait as a biometric, via phase weighted magnitude spectra", Proceedings of the 1st International Conference on Audio- and Video-Based Biometric Person Authentication, pp. 95-102, 1997.
[2] L. Lee and W.E.L. Grimson, "Gait analysis for recognition and classification", Proc. IEEE Int. Conf. Automatic Face and Gesture Recognition, pp. 148-155, Washington DC, May 2002.
[3] D.K. Wagg and M.S. Nixon, "On automated model-based extraction and analysis of gait", Proc. of 6th IEEE International Conference on Automatic Face and Gesture Recognition, pp. 11-16. Seoul, Korea, May 2004.
[4] A.Y. Johnson and A.F. Bobick, "A multi-view method for gait recognition using static body parameters", Proc. of 3rd International Conference on Audio and Video Based Biometric Person Authentication, pp. 30-311, Halmstad, Sweden, June 2001.
[5] R. Urtasun and P. Fua, "3D tracking for gait characterization and recognition", Proc. of 6th IEEE international Conference on Automatic Face and Gesture Recognition, pp. 17-22, Seoul, Korea, May 2004.
[6] R. Zhang, C. Vogler, and D. Metaxas, "Human gait recognition at sagittal plane", Image Vision Computing, 25(3), pp. 321-330, 2007.
[7] J.J. Little and J.E. Boyd, "Recognizing People by Their Gait: The Shape of Motion", Videre: J. Computer Vision Research, vol. 1, no. 2, pp. 1-32, 1998.
[8] P.S. Huang, C.J. Harris, M.S. Nixon, "Recognising humans by gait via parametric canonical space", Artificial Intelligence in Engineering 13, pp. 359-366, 1999.
[9] L. Wang, T. Tan, H. Ning, and W. Hu, "Silhouette Analysis-Based Gait Recognition for Human Identification", IEEE T. on PAMI, vol. 25, no. 12, pp. 1505-1518, 2003.
[10] J. Han, B. Bhanu, "Individual Recognition Using Gait Energy Image", IEEE T. on PAMI, vol. 28, no. 2, pp. 316-322, 2006.
[11] X. Yang, Y. Zhou, T. Zhang, G. Shu, J. Yang, "Gait recognition based on dynamic region analysis", Signal Processing 88, pp. 2350-2356, 2008.
[12] J. Liu, N. Zheng, "Gait history image: a novel temporal template for gait recognition", IEEE ICME, Beijing, China, July 2007.
[13] E. Zhang, Y. Zhao, W. Xiong, "Active energy image plus 2DLPP for gait recognition", Signal Processing, Volume 90, Issue 7, pp. 2295-2302, July 2010.
[14] T. H.W. Lam, K.H. Cheung, and J. N.K. Liu, "Gait flow image: A silhouette-based gait representation for human identification", Pattern Recognition, vol.44, pp. 973-987, 2011.
[15] Q. Ma, S. Wang, D. Nie, and J. Qiu, "Recognizing Humans Based on Gait Moment Image", in IEEE ACIS, 2007.
[16] X. Yang, Y. Zhou, T. Zhang, E. Zheng, and J. Yang, "Gabor phase based gait recognition", Electronics Letters, vol. 44 no. 10, pp. 620-621, May 2008.
[17] C. Chen, J. Liang, H. Zhao, H. Hu, and J. Tian, "Frame difference energy image for gait recognition with incomplete silhouettes", Pattern Recognition Letters 30, pp. 977-984, 2009.
[18] H. Zhang, Z. Liu, H. Zhao, "Gait Modeling and Identifying Based on Dynamic Template Matching", Journal of Computational Information Systems, vol.7, no.4, pp. 1155-1162, 2011.
[19] A. Kale, A. Sundaresan, A.N. Rajagopalan, N.P. Cuntoor, A.K. Roy-Chowdhury, V. Krüger, and R. Chellappa, "Identification of Humans Using Gait", IEEE T. on IP, vol. 13, no. 9, pp. 1163-1173, 2004.
[20] M.H. Cheng, M.F. Ho, C.L. Huang, "Gait analysis for human identification through manifold learning and HMM", Pattern Recognition, 41, pp. 2541-2553, 2008.
[21] S. Sarkar, P.J. Phillips, Z. Liu, I.R. Vega, P. Grother, and K.W. Bowyer, "The human ID gait challenge problem: datasets, performance, and analysis", IEEE T. on PAMI, 27, pp. 162-177, 2005.
[22] J.D. Shutler and M.G. Grant, "The large Southampton HiD human gait database", Technical report, ISIS, University of Southampton, U.K., 2002.
[23] S. Yu, D. Tan, and T. Tan, "A framework for evaluating the effect of view angle, clothing and carrying condition on gait recognition", ICPR, 2006.
[24] E. Xing, A. Ng, and M. Jordan, "Distance metric learning with application to clustering with side-information", NIPS, 2002.
[25] A. Bar-Hillel, T. Hertz, N. Shental, and D. Weinshall, "Learning a Mahalanobis Metric from Equivalence Constraints", Journal of Machine Learning Research, 2005.
[26] J. Goldberger, S. Roweis, G. Hinton, and R. Salakhutdinov, "Neighbourhood components analysis", NIPS, 2005.
[27] K. Weinberger, J. Blitzer, and L. Saul, "Distance metric learning for large margin nearest neighbor classification", NIPS, 2006.
[28] P. Mudigonda, P. Torr, and A. Zisserman, "Invariant Large Margin Nearest Neighbor Classifier", ICCV, 2007.
[29] P.N. Belhumeur, J.P. Hespanha, D.J. Kriegman, "Eigenfaces vs. Fisherfaces: Recognition Using Class Specific Linear Projection", IEEE T. on PAMI, vol.19 no.7, pp. 711-720, 1997.
[30] X. He and P. Niyogi, "Locality preserving projections", Proceedings of the Conference Advances in Neural Information Processing Systems, 2003.
[31] X. He, S. Yan, Y. Hu, P. Niyogi, and H. Zhang, "Face Recognition Using Laplacianfaces", IEEE T. on PAMI, vol. 27, no. 3, pp. 328-340, 2005.
[32] S.T. Roweis and L.K. Saul, "Nonlinear Dimensionality Reduction by Locally Linear Embedding", Science, vol. 290, 2000.
[33] L.K. Saul and S.T. Roweis, "Think Globally, Fit Locally: Unsupervised Learning of Low Dimensional Manifolds", J. Machine Learning Research, vol. 4, pp. 119-155, 2003.
[34] D. Cai, X. He, J. Han, and H.J. Zhang, "Orthogonal laplacianfaces for face recognition", IEEE T. on IP, 15(11), pp. 3608-3614, 2006.
[35] J. Makihara, R. Sagawa, and Y. Mukaigawa, "Gait recognition using a view transformation model in the frequency domain", ECCV, 2006.
[36] W. Kusakunniran, Q. Wu, H. Li, and J. Zhang, "Multiple view gait recognition using view transformation model based on optimized gait energy images", ICCV workshops, 2009.
[37] J. Zhang, K. Huang, R. He, and T. Tan, "Robust view transformation model for gait recognition", ICIP, 2011
[38] W. Kusakunniran, Q. Wu, J. Zhang, and H. Li, "Gait recognition under various viewing angles based on correlated motion regression", IEEE T. on CSVT, 22(6), pp. 966-980, 2012.
[39] W. Kusakunniran, Q. Wu, J. Zhang, and H. Li, "Cross-view and multi-view gait recognitions based on view transformation model using multi-layer perceptron", Pattern Recognition Letters, vol.33, no.7, pp. 882-889, 2012.
[40] J.B. Tenenbaum and W.T. Freeman, "Separating style and content with bilinear models", Neural Computation, vol.12, issue 6, pp. 1247-1283, 2000.
[41] http://marathon.csee.usf.edu/GaitBaseline/
[42] http://www.gait.ecs.soton.ac.uk/
[43] http://www.cbsr.ia.ac.cn/english/Gait%20Databases.asp
[44] G. Zhao, G. Liu,H. Li, and M. Pietikainen, "3D gait recognition using multiple cameras", 7th International Conf. on Automatic Face and Gesture Recognition, pp. 529-534, April 2006.
[45] B. Bhanu and J. Han, "Human recognition at a distance in video", Series: Advances in Computer vision and Pattern Recognition, Springer, pp. 65-94, 2010.

指導教授

范國清(Kuo-chin Fan)

審核日期

2013-7-15

推文