請用此 Handle URI 來引用此文件:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/56754
完整後設資料紀錄
DC 欄位 | 值 | 語言 |
---|---|---|
dc.contributor.advisor | 王傑智(Chieh-Chih Wang) | |
dc.contributor.author | Yi Chiang | en |
dc.contributor.author | 江懿 | zh_TW |
dc.date.accessioned | 2021-06-16T05:46:23Z | - |
dc.date.available | 2014-08-16 | |
dc.date.copyright | 2014-08-16 | |
dc.date.issued | 2014 | |
dc.date.submitted | 2014-08-11 | |
dc.identifier.citation | [1] Z.-Q. Luo and P. Tseng, “On the convergence of the coordinate descent method for
convex differentiable minimization,” Journal of Optimization Theory and Applica- tions, 1992. [2] R. Pardo, The evaluation and optimization of trading strategies, vol. 314. John Wiley & Sons, 2011. [3] R. H. Chung, F. Y. Chin, K.-Y. K. Wong, K. Chow, T. Luo, and H. S. Fung, “Effi- cient block-based motion segmentation method using motion vector consistency.,” Conference on Machine Vision Applications (IAPR-MVA), 2005. [4] Y. Wang and S. Huang, “An efficient motion segmentation algorithm for multibody rgb-d slam,” Australasian Conference on Robotics and Automation (ACRA), 2013. [5] S. Perera and N. Barnes, “Maximal cliques based rigid body motion segmentation with a rgb-d camera,” Asian Conference on Computer Vision (ACCV), 2012. [6] K. Schindler, “Spatially consistent 3d motion segmentation,” International Confer- ence on Image Processing (ICIP), 2005. [7] M. Van den Bergh and L. Van Gool, “Real-time stereo and flow-based video segmen- tation with superpixels,” Workshop on Applications of Computer Vision (WACV), 2012. [8] M. Narayana, A. Hanson, and E. Learned-Miller, “Coherent motion segmentation in moving camera videos using optical flow orientations,” International Conference on Computer Vision (ICCV), 2013. [9] E. Herbst, X. Ren, and D. Fox, “Object segmentation from motion with dense feature matching,” International Conference on Robotics and Automation (ICRA), 2012. [10] J. Stückler and S. Behnke, “Efficient dense 3d rigid-body motion segmentation in rgb-d video,” British Machine Vision Conference (BMVC), 2013. [11] G. Rosman, A. M. Bronstein, M. M. Bronstein, and R. Kimmel, “Articulated motion segmentation of point clouds by group-valued regularization,” Eurographics Work- shop on 3D Object Retrieval (EG 3DOR), 2012. [12] S. Hadfield and R. Bowden, “Go with the flow: hand trajectories in 3d via clustered scene flow,” International Conference on Image Analysis and Recognition (ICIAR), 2012. [13] M. Patriksson, “Decomposition methods for differentiable optimization problems over cartesian product sets,” Computational optimization and applications, 1998. [14] L. Grippo and M. Sciandrone, “On the convergence of the block nonlinear gauss– seidel method under convex constraints,” Operations Research Letters, 2000. [15] S. Rusinkiewicz and M. Levoy, “Efficient variants of the icp algorithm,” 3D Digital Imaging and Modeling (3DIM), 2001. [16] K. S. Arun, T. S. Huang, and S. D. Blostein, “Least-squares fitting of two 3-d point sets,” The IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 1987. [17] B. K. Horn, “Closed-form solution of absolute orientation using unit quaternions,” Journal of the Optical Society of America A (JOSA A), 1987. [18] M. W. Walker, L. Shao, and R. A. Volz, “Estimating 3-d location parameters using dual number quaternions,” Conference on Computer Vision, Graphics, and Image Processing (CVGIP), 1991. [19] C. V. Nguyen, S. Izadi, and D. Lovell, “Modeling kinect sensor noise for improved 3d reconstruction and tracking,” 3D Imaging, Modeling, Processing, Visualization and Transmission (3DIMPVT), 2012. [20] C. Zhang and Z. Zhang, “Calibration between depth and color sensors for commodity depth cameras,” International Conference on Multimedia and Expo (ICME), 2011. | |
dc.identifier.uri | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/56754 | - |
dc.description.abstract | 稠密 RGB-D 運動圖像分割是計算機視覺、影像處理與機器人學中重要的前處理模塊。本論文提出了一個基於深度資訊且不依賴於顏色的運動圖像分割演算法。此演算法以最佳化框架將有著一致運動的物件從背景與彼此分割並計算其運動參數。本方法在運動分割的同時也計算稠密點對應。本論文並提出了一個基於區塊約束非線性高斯 -賽得爾迭代 [1] 與優先步驟搜尋法 [2] 的數值方法來有效率的解決所提出的最佳化問題。本數值方法將變數分類並決定其優化的順序並對演算法的收歛性加以證明。此演算法在移動相機且佈滿非剛體動態物件的環境下有
良好的表現。 | zh_TW |
dc.description.abstract | Dense RGB-D video motion segmentation is an important preprocessing module in computer vision, image processing and robotics. A motion segmentation algorithm based on an optimization framework which utilizes depth information only is presented in this thesis. The proposed optimization framework segments and estimates rigid motion parameters of each locally rigid moving objects with coherent motion. The proposed method also calculates dense point correspondences while performing segmentation. An efficient numerical algorithm based on Constrained Block Nonlinear Gauss-Seidel (CNLGS) algorithm [1] and Prioritized Step Search [2] is proposed to solve the optimization problem. It classifies variables including point correspondences into groups and determines the ordering of variables to optimize. We prove the proposed numerical algorithm to converge to a theoretical
bound. The proposed algorithm works well with a moving camera in highly dynamic urban scenes with non-rigid moving objects. | en |
dc.description.provenance | Made available in DSpace on 2021-06-16T05:46:23Z (GMT). No. of bitstreams: 1 ntu-103-R01922037-1.pdf: 26760313 bytes, checksum: ee535b0aa5e7e94925d91876a2bf16c2 (MD5) Previous issue date: 2014 | en |
dc.description.tableofcontents | ABSTRACT iii
TABLE OF CONTENTS iv List of Figures vi 1 Introduction 1 2 Related Work 5 3 Approach 7 3.1 Optimization Framework . . . . . . . . . . . . . . . . . . . . . . . . . . 7 3.2 The Constrained Block Nonlinear Gauss-Seidel Algorithm . . . . . . . . 9 3.3 The Prioritized Step Search . . . . . . . . . . . . . . . . . . . . . . . . . 13 3.4 Motion Assignment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 3.5 Motion Estimation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 4 System Flow 18 4.1 Preprocessing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 4.2 Correspondence Finding . . . . . . . . . . . . . . . . . . . . . . . . . . 20 4.3 Motion Parameter Estimation . . . . . . . . . . . . . . . . . . . . . . . . 21 4.3.1 Inlier Filtering . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 4.4 Motion Assignment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 4.5 Merge and Refine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 5 Experiment Results 25 5.1 Experiment Setup and Discussion . . . . . . . . . . . . . . . . . . . . . 25 5.2 The Night Market Motion Segmentation Benchmark . . . . . . . . . . . 27 6 Conclusion and Future Work 33 Bibliography 34 | |
dc.language.iso | en | |
dc.title | 以優先步驟高斯-賽得爾法完成基於移動深度相機在擁擠都市環境之稠密對應點估測與運動分割 | zh_TW |
dc.title | A Prioritized Gauss-Seidel Method for Dense Correspondence Estimation and Motion Segmentation in Crowded Urban Areas with a Moving Depth Camera | en |
dc.type | Thesis | |
dc.date.schoolyear | 102-2 | |
dc.description.degree | 碩士 | |
dc.contributor.oralexamcommittee | 鍾聖倫(Sheng-Luen Chung),傅立成(Li-Chen Fu),莊永裕(Yung-Yu Chuang),陳祝嵩(Chu-song Chen) | |
dc.subject.keyword | 對應,分割,深度,相機,高斯,賽得爾, | zh_TW |
dc.subject.keyword | correspondence,segmentation,depth,camera,Gauss,Seidel, | en |
dc.relation.page | 36 | |
dc.rights.note | 有償授權 | |
dc.date.accepted | 2014-08-11 | |
dc.contributor.author-college | 電機資訊學院 | zh_TW |
dc.contributor.author-dept | 資訊工程學研究所 | zh_TW |
顯示於系所單位: | 資訊工程學系 |
文件中的檔案:
檔案 | 大小 | 格式 | |
---|---|---|---|
ntu-103-1.pdf 目前未授權公開取用 | 26.13 MB | Adobe PDF |
系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。