摘要
视觉测量是计算机视觉研究中的核心问题之一。本论文以单目序列图像为对象，以对空飞行器的被动检测与参数测量需求为依托，针对目前单目视觉测量中的两个主要难点问题：(1)投影方程的数学反演，(2)从输入图像到目标三维模型的2D-3D特征投影对应关系建立，进行了重点研究和探索：
1）针对当前P3P问题的闭式解法仍存在数值精度低、数值稳定性差，而迭代解法无法获取全部可行解的问题，提出一种具有高数值精度和数值稳定性的P3P问题半闭式解法。该方法利用多个可行解之间的“共线约束”关系，将待解方程组阶数降低至最低的2阶，实现了在获取P3P问题全部可行解同时，保持与迭代解法相当的数值精度。
2）在未给定任何先验特征投影对应关系情况下的三维刚体目标姿态估计中，针对现有迭代估计方法仍存在收敛半径小和收敛速度慢的问题，提出了一种新的基于2D-3D泛轮廓点对应的迭代姿态估计方法。该方法从显式建立由输入图像到目标三维模型在泛轮廓点上的2D-3D投影对应关系出发，通过迭代同步完成了特征投影对应关系的确定和目标三维姿态参数的估计，显著改进了现有方法的收敛半径和收敛速度。
3）在前述迭代姿态估计算法框架的基础上，提出了一种基于非欧多特征距离图的三维刚体目标姿态估计方法。该方法从如何在迭代过程中，进一步提高泛轮廓点上、由输入图像到目标三维模型的2D-3D特征投影对应关系准确性出发，提出了非欧氏、多特征距离图的概念，在不显著增加计算量的前提下有效增大了迭代姿态估计算法的收敛半径。
4）针对单目视觉测量的实时处理需求，提出了一种基于多尺度空间、精度可伸缩的快速三维刚体目标姿态估计方法，在保证参数测量精度的前提下，有效降低了三维刚体目标姿态估计的算法复杂度。
5）针对迭代姿态估计方法所依赖的目标轮廓提取问题，从准确性和处理速度两个不同角度，提出一种具有高噪声鲁棒性的改进型Chan-Vese主动轮廓模型和一种基于离散水平集表示的快速主动轮廓演进方案，保证了强噪声下边缘收敛的准确性，并大幅度提高了边缘分割速度。
Vision-based metrology is one of the essential problems in computer vision.Supported by the actual application demand for monocular vision based passivedetection and parameter measurement of airborne targets, this dissertation isdedicated to the research of the two major problems in monocular vision basedmetrology:(1) the problem of mathematical inverse estimation of camera’sprojection relationship equation;(2) the problem of establishing the2D-3D featureprojection correspondence from the input image to object’s3D model, arethoroughly studied:
1) For P3P problem, to solve the problem that the numerical precision andstability of state-of-art closed-form methods are still relative low meanwhileiterative methods can return only one of the multiple feasible solutions, a newsemi-closed method which has high numerical precision and stability is proposed.The new method breaks through from the perspective of lowering the order of thepolynomial equations to be solved, and decreases the order to the ever possiblelowest2by exploiting the “collinear constraint” of all the feasible solutions of P3Pproblem. With the new semi-closed method, all the feasible solutions of P3Pproblem can be retrieved with comparable numerical precision to iterative methods.
2) For the pose estimation of3D rigid object when there is no2D-3D featureprojection correspondence from the input image to object’s3D model given apriori, aiming at the insufficiency of state-of-art methods in convergence radiusand speed, a new iterative pose estimation method based on general2D-3D contourpoint correspondence is proposed. The new method makes the breakthrough fromthe perspective of explicitly establishing the2D-3D feature projectioncorrespondence between the input image and object’s3D model on general contourpoints, and solving the feature correspondence establishing task and the3D poseestimation task iteratively and simultaneously. Evident improvements are gained onconvergence radius and speed performance.
3) Based on the above iterative pose estimation algorithm framework, a newnon-Euclidean multi-feature distance map based3D rigid object pose estimationmethod is proposed. The proposed method concentrates on increasing the accuracyof the2D-3D contour points projection correspondence establishing from the inputimage to object’s3D model, and by introducing the concept of non-Euclideanmulti-feature distance map, further improvement on pose estimation convergenceradius is obtained, while without incurring too much additional computationalburden.
4) Targeting at the real-time processing requirement of monocular visionmetrology task, a multi-scale space based fast pose estimation method for3D rigidobject is proposed. The new method effectively reduces the computationalcomplexity of the monocular vision based pose estimation system without losingthe required parameter estimation precision, and has been successfully applied tovideo-oriented real-time3D rigid object pose estimation.
5) For the object contour extraction problem on which the above proposediterative pose estimation methods rely, a new noise-robust modified Chan-Veseactive contour model and a fast active contour implementation scheme based ondiscretized level set representation are proposed, from the perspective ofsegmentation accuracy and processing speed respectively. The former can handlesevere noise while guarantee the active contour to converge to target edges moreprecisely. The latter improves on segmentation speed performance by a largemargin.
引文
[1] Hartley R., Zisserman A. Multiple view geometry in computer vision,2nd ed., CambridgeUniversity Press,2003
[2] Bishop C. M. Pattern recognition and machine learning, Springer,2006
[3] Xu G. Image Based Modeling and Rendering, Wuhan University Press,2006
[4] Zeng H. Study on Visual Metrology and Pose Determination, Phd dissertation, Institue ofAutomation, Chinese Academic of Science,2007
[5] Wang L. Epipolar geometry estimation and camera calibration, Phd dissertation, Institue ofAutomation, Chinese Academic of Science,2007
[6] Horprasert T., Yacoob Y., Davis L. S. Computing3-D head orientation from a monocular imagesequence, Proceedings of the Second International Conference on Automatic Face and GestureRecognition,1996, pp.242-247
[7] Yao P. P., Evans G., Calway A. Using affine correspondence to estimate3-D facial pose, ICIP2001, Vol.3, pp.919-922
[8] Liang G. Y., Zha H. B., Liu H. Affine correspondence based head pose estimation for a sequenceof images by using a3D model, Proceedings of the Sixth IEEE international conference onAutomatic face and gesture recognition,2004, pp.481-484
[9] DeMenthon D., Davis L. S. Model-based object pose in25lines of code, IJCV, Vol.15, No.1-2,1995, pp.123-141
[10] David P., DeMenthon D. F., Duraiswami R., Samet H. SoftPOSIT: simultaneous pose andcorrespondence determination, IJCV, Vol.59, No.3,2004, pp.259-284
[11] David P., DeMenthon D., Duraiswami R., Samet H. Simultaneous pose and correspondencedetermination using line features, CVPR2003, Vol.2, pp.424-431
[12] Criminisi A., Reid I., Zisserman A. Single view metrology, IJCV, Vol.40, No.2,2000, pp.123-148
[13] Zhang Z. Y. Flexible camera calibration by viewing a plane from unknown orientations, ICCV99,Vol.1, pp.666-673
[14] Zhang Z. Y. A flexible new technique for camera calibration, IEEE Trans. on IntelligentTransportation Systems, Vol.22, No.11,2000, pp.1330-1334
[15] Clarke T. A., Fryer J. G. The development of camera calibration methods and models,Photogrammetric Record, Vol.16, No.91,1998, pp.51-66
[16] Tsai R. Y. A versatile camera calibration technique for high accuracy3D machine visionmetrology using off-the-shelf TV cameras and lenses, IEEE Journal of Robotics and Automation,Vol. RA-3, No.4,1987, pp.323-344
[17] Grunert J. A. Das Pothenotische Problem in erweiterter Gestalt nebst über seine Anwendungen inder Geod sie, Grunerts Archiv für Mathematik und Physik, Band1,1841, pp.238-248
[18] Fischler M. A., Bolles R. C. Random Sample Consensus: a paradigm for model fitting withapplication to image analysis and automated cartography, ACM Graphics and Image Processing,Vol.24, No.6,1981, pp.381-395
[19] Hu Z. Y., Wu F. C. A note on the number of solution of the non-coplanar P4P problem, IEEETrans. on Pattern Analysis and Machine Intelligence, Vol.24, No.4,2002, pp.550-555
[20] Wu Y. H., Hu Z. Y. PnP problem revisited, Journal of Mathematical Imaging and Vision, Vol.24,No.1,2006, pp.131-141
[21] Lu C. P., Hager G. D., Mjolsness E. Fast and globally convergent pose estimation from videoimages, IEEE Trans. on Pattern Analysis and Machine Intelligence, Vol.22, No.6,2000, pp.610-622
[22] Leng D. W., Sun W. D. Finding all the solutions of PnP problem, IST2009, IEEE, Shenzhen, pp.348-352
[23] Noguer F. M., Lepetit V., Fua P. Accurate non-iterative O(n) solution to the PnP problem, ICCV2007, IEEE, Rio de Janeiro, pp.2168-2175
[24] Gao X. S., Hou X. R., Tang J. L., Cheng H. F. Complete solution classification for theperspective-three-point problem, IEEE Trans. on Pattern Analysis and Machine Intelligence, Vol.25, No.8,2003, pp.930-943
[25] Yang L. A simplified algorithm for solution classification of the perspective-three-point problem,MMRC, Academia Sinica, No.17,1998, pp.135-145
[26] Burschka D., Mair E. Direct pose estimation with a monocular camera, Lecture Notes inComputer Science, Robot Vision,2008, Vol.4931, pp.440-453
[27] Merritt E. L. Explicit three-point resection in space, Photogrammetric Engineering, Vol. XV, No.4,1949, pp.649-655
[28] Haralick R. M., Lee C., Ottenberg K., Nolle M. Review and analysis of solution of the three pointperspective pose estimation problem, IJCV, Vol.13, No.3,1994, pp.331-356
[29] DeMenthon D., Davis L. S. Exact and approximate solutions of the Perspective-Three-Pointproblem, IEEE Trans. on Pattern Analysis and Machine Intelligence, Vol.14, No.11,1992, pp.1100-1105
[30] Quan L., Lan Z. Linear N-point camera pose determination, IEEE Trans. on Pattern Analysis andMachine Intelligence, Vol.21, No.8,1999, pp.744-780
[31] Fiore P. D. Efficient linear solution of exterior orientation, IEEE Trans. on Pattern Analysis andMachine Intelligence, Vol.23, No.2,2001, pp.140-148
[32] Horaud R., Conio B., Leboulleux O. An analytic solution for the Perspective4-Point problem,Computer Vision Graphics and Image Processing, Vol.47, No.1,1989, pp.33-44
[33] Haralick R. M. Pose estimation from corresponding point data, IEEE Trans. on Systems, Man andCybernetics, Vol.19, No.6,1989, pp.1426-1446
[34] Triggs B. Camera pose and calibration from4or5known3D points, ICCV99, pp.278-284
[35] Ansar A., Daniilidis K. Linear pose estimation from points or lines, IEEE Trans. on PatternAnalysis and Machine Intelligence, Vol.25, No.5,2003, pp.578-589
[36] Christy S., Horaud R. Iterative pose computation from line correspondences, CVIU, Vol.73, No.1,1999, pp.137-144
[37] Liu Y., Huang T. S., Faugeras O. D. Determination of camera location from2-D to3-D line andpoint correspondences, IEEE Trans. on Pattern Analysis and Machine Intelligence, Vol.12, No.1,1990, pp.28-37
[38] Navab N., Faugeras O. D. Monocular pose determination from lines: critical sets and maximumnumber of solutions, CVPR93, pp.254-260
[39] Horaud R. New methods for matching3D objects with single perspective views, IEEE Trans. onPattern Analysis and Machine Intelligence, Vol.9, No.3,1987, pp.401-412
[40] Dhome M., Richetin M., Lapresté J. T., Rives G. Determination of the attitude of3D objects froma single perspective view, IEEE Trans. on Pattern Analysis and Machine Intelligence, Vol.11, No.12,1989, pp.1265-1278
[41] Hmam H., Kim J. Aircraft recognition and pose estimation, International Conference on VisualCommunications and Image Processing,2000, pp.1198-1209
[42] Schweighofer G., Pinz A. Robust pose estimation from a planar target, IEEE Trans. on PatternAnalysis and Machine Intelligence, Vol.28, No.12,2006, pp.2024-2030
[43] Hanning T., Schoene R., Graf S. A closed form solution for monocular re-projective3D poseestimation of regular planar patterns, ICIP2006, Vol.1-7, pp.2197-2200
[44] McInroy J. E., Qi Z. A novel pose estimation algorithm based on points to regionscorrespondence, Journal of Mathematical Imaging and Vision, Vol.30, No.2,2008, pp.195-207
[45] Basri R., Jacobs D. W. Projective alignment with regions, IEEE Trans. on Pattern Analysis andMachine Intelligence, Vol.23, No.5,2001, pp.519-527
[46] Jacobs D., Basri R.3D to2D pose determination with regions, IJCV, Vol.34, No.2-3,1999, pp.123-145
[47] Tahri O., Chaumette F. Complex objects pose estimation based on image moment invariants, Proc.IEEE International Conference on Robotics and Automation, Barcelona, Spain,2005, pp.436-441
[48] Kyriakoulis N., Gasteratos A. Color-based monocular visuoinertial3-D pose estimation of aVolant robot, IEEE Trans. on Instrumentation and Measurement, Vol.59, No.10,2010, pp.2706-2715
[49] Viksten F., Forssén P. E., Johansson B., Moe A. Comparison of local image descriptors for full6degree-of-freedom pose estimation, IEEE International Conference on Robotics and Automation,Kobe, Japan,2009, pp.1139-1146
[50] Shan G. L., Ji B., Zhou Y. F. A review of3D pose estimation from a monocular image sequence,CISP2009, Tianjin, pp.1-5
[51] Douxchamps D., Chihara K. High-accuracy and robust localization of large control markers forgeometric camera calibration, IEEE Trans. On Pattern Analysis and Machine Intelligence, Vol.31,No.2,2009, pp.376-383
[52] Azuma R. T. A survey of augmented reality, Presence: Teleoperators and Virtual Environments,Vol.6, No.4,1997, pp.355-385
[53] Zorpette G. Weta Digital reverse engineers the human face, IEEE Spectrum, Vol.47, No.4,2010,pp.18
[54] Lepetit V., Fua P. Monocular model-based3D tracking of rigid objects: a survey, Foundationsand Trends in Computer Graphics and Vision, Vol.1, No.1,2005, pp.1-89
[55] Lowe D. G. Distinctive image features from scale-invariant keypoints, IJCV, Vol.60, No.2,2004,pp.91-110
[56] Ozuysal M., Fua P., Lepetit V. Fast keypoint recognition in ten lines of code, CVPR,2007, pp.1-8
[57] Schmid C., Mohr R., Bauckhage C. Evaluation of interest point detector, IJCV, Vol.37, No.2,2000, pp.151-172
[58] Bay H., Tuytelaars T., Gool L. V., Zurich E. SURF: speeded up robust features, CVIU, Vol.110,No.3,2008, pp.346-359
[59] Smith S. M., Brady J. M. SUSAN—a new approach to low level image processing, IJCV, Vol.23,No.1,1997, pp.45–78
[60] Harris C., Stephens M. A combined corner and edge detector, Proceedings of the4th AlveyVision Conference,1988, pp.147-151
[61] Kovesi P. Phase congruency detects corners and edges, The Australian Pattern RecognitionSociety Conference,2003, pp.309-318
[62] Mikolajczyk K., Tuytelaars T., Schmid C., Zisserman A., Matas J., Schaffalitzky F., Kadir T.,Gool L. V. A comparison of affine region detectors, IJCV, Vol.65, No.2,2005, pp.43-72
[63] Donoser M., Bischof H. Efficient maximally stable extremal region (MSER) tracking, CVPR2006, Vol.1, pp.553-560
[64] Dunker J., Hartmann G., St hr M. Single view recognition and pose estimation of3D objectsusing sets of prototypical views and spatially tolerant contour representations, ICPR96, Vol.4, pp.14-18
[65] Poggio T., Edelman S. A network that learns to recognize three-dimensional objects, Nature, Vol.343,1990, pp.263-266
[66] Iwashita Y., Kurazume R., Konishi K., Nakamoto M., Hashizume M., Hasegawa T. Fastalignment of3D geometrical models and2D grayscale images using2D distance maps, Systemsand Computers in Japan, Vol.38, No.14,2007, pp.1889–1899
[67] Dambrevile S., Sandhu R., Yezzi A., Tannenbaum A. A geometric approach to joint2Dregion-based segmentation and3D pose estimation using a3D shape prior, SIAM Journal onImaging Sciences, Vol.3, No.1,2010, pp.110-132
[68] Cui Y., Hildenbrand D. Pose estimation based on geometric algebra, GraVisMa2009, pp.17-24
[69] Osher S., Fedkiw R. Level set methods and dynamic implicit surfaces, Springer,2003
[70] Sethian J. A. Level set methods and fast marching methods,2nded., Cambridge University Press,1999
[71] Chan T. F., Vese L. A. Active contours without edges, IEEE Trans. on Image Processing, Vol.10,No.2,2001, pp.266-277
[72] Li C., Xu C., Gui C., Fox M. D. Level set evolution without re-initialization: a new variationalformulation, CVPR2005, Vol.1, pp.430-436
[73] Zhu S. C., Yuille A., Lee T. S. Region competition: Unifying snakes, region growing andBayes/MDL for multi-band image segmentation, ICCV95, pp.416–423
[74] Caselles V., Kimmel R., Sapiro G. Geodesic active contours, IJCV, Vol.22, No.1,1997, pp.61–79
[75] Appleton B., Talbot H. Globally optimal geodesic active contours, Journal of MathematicalImaging and Vision, Vol.23, No.1,2005, pp.67-86
[76] Li H., Yezzi A. Local or global minima: flexible dual-front active contours, IEEE Trans. OnPattern Analysis and Machine Intelligence, Vol.29, No.1,2007, pp.1-14
[77] Mao H., Liu H., Shi P. Neighbor-constrained active contours without edges, CVPR2008, pp.1-7
[78] Mishra A. K., Fieguth P. W., Clausi A. Decoupled active contour (DAC) for boundary detection,IEEE Trans. On Pattern Analysis and Machine Intelligence, Vol.33, No.2,2011, pp.310-324
[79] Zhao H. K., Chan T., Merriman B., Osher S. A variational level set approach to multiphasemotion, Journal of Computational Physics, Vol.127, No.1,1996, pp.179-195
[80] Goldenberg R., Kimmel R., Rivlin E., Rudzsky M.,“Fast geodesic active contours, IEEE Trans.on Image Processing, Vol.10, No.10,2001, pp.1467-1474
[81] Bresson X., Esedoglu S., Vandergheynst P., JThiran. P., Osher S. Fast global minimization of theactive contour/snake model, Journal of Mathematical Imaging and Vision, Vol.28, No.2,2007,pp.151-167
[82] Li C., Kao C. Y., Gore J. C., Ding Z. Minimization of region-scalable fitting energy for imagesegmentation, IEEE Trans. on Image Processing, Vol.17, No.10,2008, pp.1940-1949
[83] Vasilevskiy A., Siddiqi K. Flux-maximizing geometric flows, IEEE Trans. on Pattern Analysisand Machine Intelligence, Vol.24, No.12,2002, pp.1565-1578
[84] Paragios N., Deriche R. Geodesic active regions and level set methods for supervised texturesegmentation, IJCV, Vol.46, No.3,2002, pp.223–247
[85] Tsai A., Yezzi A., Willsky A. S. Curve evolution implementation of the Mumford–Shahfunctional for image segmentation, denoising, interpolation, and magnification, IEEE Trans. onImage Processing, Vol.10, No.8,2001, pp.1169–1186
[86] Chan T. F., Sandberg B. Y., Vese L. A. Active contours without edges for vector-valued images,Journal of Visual Communication and Image Representation, Vol.11, No.2,2000, pp.130-141
[87] Chan T. F., Shen J. H. Image processing and analysis: variational, PDE, wavelet, and stochasticmethods, SIAM,2005
[88] Perona P., Malik J. Scale-space and edge detection using anisotropic diffusion, IEEE Trans. onPattern Analysis and Machine Intelligence, Vol.12, No.7,1990, pp.629–639
[89] Weikert J. Anisotropic diffusion in image processing, ECMI series, Teubner-Verlog, Germany,1998
[90] Perona P. Orientation diffusions, IEEE Trans. on Image Processing, Vol.7, No.31998, pp.457–467
[91] Alvarez L., Mazorra L. Signal and image restoration using shock filters and anisotropic diffusion,SIAM Journal on Numerical Analysis, Vol.31, No.2,1994, pp.590-605
[92] Guichard F., Morel J. Geometric partial differential equations and iterative filtering, Proceedingsof the Fourth International Symposium on Mathematical Morphology and Its Applications toImage and Signal Processing,1998, pp.127-138
[93] Osher S., Rudin L. I. Feature-oriented image enhancement using shock filters, SIAM Journal onNumerical Analysis, Vol.27, No.4,1990, pp.919-940
[94] Kass M., Witkin A., Terzopoulos D. Snakes: active contour models, IJCV, Vol.1, No.4,1988, pp.321–331
[95] Kichenassamy S., Kumar A., Olver P., Tannenbaum A., Yezzy A. Gradient flows and geometricactive contour models, ICCV95, Cambridge, MA, pp.810–815
[96] Russo G., Smereka P. A remark on computing distance functions, Journal of ComputationalPhysics, Vol.163, No.1,2000, pp.51-67
[97] Rodrigues I., Sanches J., Dias J. B. Denoising of medical images corrupted by Poisson noise, ICIP2008, pp.1756-1759
[98] Bovik A. C. Handbook of image and video processing, Academic Press,2005
[99] Felzenszwalb P. F., Huttenlocher D. P. Distance transforms of sampled functions, CornellComputing and Information Science TR2004-1963,2004
[100] Gomez P., Hernandez J., Lopez J. On the reinitialization procedure in a narrow-band locallyrefined level set method for interfacial flows, International Journal for Numerical Methods inEngineering, Vol.63, No.10,2005, pp.1478-1512
[101] Zhao H. Parallel implementations of the fast sweeping method, Journal of ComputationalMathematics, Vol.25, No.4,2007, pp.421-429
[102] Wu Z. L., Wang Z. L., Luo L. M. Fast level set image segmentation on graphics processing unit,Journal of Image and Graphics, Vol.9, No.6,2004, pp.679-683
[103] Zhou X. The video object tracking system based on active contour model, Master Thesis,Huadong Normal University,2006
[104] Wang D. Object segmentation based on level set and an ATR system, Master Thesis, HuazhongUniversity of Science and Technology,2007
[105] Alvarez L., Guichard F., Lions P. L., Morel J. M. Axioms and fundamental equations of imageprocessing, Archive for Rational Mechanics and Analysis, Vol.123, No.3,1993, pp.199-257
[106] Wilson W. Visual servo control of robots using Kalman filter estimates of robot pose relative towork-pieces, Visual Servoing, World Scientific,1994, pp.71-104
[107] Umeyama S. Least-squares estimation of transformation parameters between two point patterns,IEEE Trans. on Pattern Analysis and Machine Intelligence, Vol.13, No.4,1991, pp.376-380
[108] Horn B. K. P., Hilden H. M., Negahdaripour S. Closed-form solution of absolute orientation usingorthonormal matrices, Journal of the Optical Society of America A, Vol.5, No.7,1988, pp.1127-1135
[109] Verschelde J. Algorithm795: PHCpack: a general-purpose solver for polynomial systems byhomotopy continuation, ACM Trans. on Mathematical Software, Vol.25, No.2,1999, pp.251-276
[110] http://www.math.uic.edu/~jan
[111] Lee T. K., Drew M. S.3D object recognition by eigen-scale-space of contours, SSVM2007, Vol.4485, pp.883-894
[112] Bertsekas D. P. Constrained optimization and Lagrange multiplier methods, Academic Press,1982
[113] Fabbri R., Costa L. D. F., Torelli J. C., Bruno O. M.2D Euclidean distance transform algorithms:a comparative survey, ACM Computing Surveys, Vol.40, No.1, Article2,2008
[114] Struik D. J. Lectures on classical differential geometry,2ndedition, Dover Publications, Inc.,1961
[115] Zollner H., Sablatnig R. Comparison of methods for geometric camera calibration using planarcalibration targets, Proceedings of the28th Workshop Austrian Association of PatternRecognition,2004, pp.234-244
[116] Kanhere N. K., Birchfield S. T. A taxonomy and analysis of camera calibration methods fortraffic monitoring applications, IEEE Trans. on Intelligent Transportation Systems, Vol.11, No.2,2010, pp.441-452
[117] Heikkil J., Silvén O. A four-step camera calibration procedure with implicit image correction,CVPR97, pp.1106-1112
[118] Grimson W. E. L., Pérez T. L., Huttenlocher D. P. Object recognition by computer: the role ofgeometric constraints, MIT Press,1990
[119] Lowe D. G. Object recognition from local scale-invariant features, ICCV99, pp.1150-1157
[120] Mikolajczyk K., Schmid C. Indexing based on scale invariant interest points, ICCV2001, pp.525-531
[121] Tieu K., Viola P. Boosting image retrieval, IJCV, Vol.56, No.1-2,2004, pp.17-36
[122] Rothganger F., Lazebnik S., Schmid C., Ponce J.3D object modeling and recognition fromphotographs and image sequences, Proceedings of Toward Category-Level Object Recognition,2006, pp.105-126
[123] Weiss I., Ray M. Model-based recognition of3D objects from single images, IEEE Trans. onPattern Analysis and Machine Intelligence, Vol.23, No.2,2001, pp.116-128
[124] Belongie S., Malik J., Puzicha J. Shape matching and object recognition using shape contexts,IEEE Trans. On Pattern Analysis and Machine Intelligence, Vol.24, No.24,2002, pp.509-522
[125] Yang M. Q., Kpalma K., Ronsin J. Shape matching and object recognition using chord contexts,International Conference Visualisation,2008,
[126] Su M. Research on multi-resolution3D recognition based human visual model, Master Thesis,Tsinghua University,2007, pp.63-69
[127] Cyr C. M., Kimia B. B. A similarity-based aspect-graph approach to3D object recognition, IJCV,Vol.57, No.1,2004, pp.5-22
[128] Chen Q., Yao J., Cham W. K.3D model-based pose invariant face recognition from multipleviews, IET Computer Vision, Vol.1, No.1,2007, pp.25-34
[129] Li S. Z., Jain A. K. Handbook of face recognition, Springer,2005
[130]洪海，姜伟，李金宝.基于单视觉原理实时提取运动物体三维信息的一种方法.中国图象图形学报，1997，第2卷第7期，pp.464-467
[131]刘昶，朱枫，欧锦军.基于门形三条直线的P3L问题的闭式解.沈阳理工大学学报，2009，第28卷第4期，pp.11-18
[132]周鑫，朱枫.关于P3P问题解的唯一性条件的几点讨论.计算机学报，2003，第26卷第12期，pp.1696-1701
[133]苏成，徐迎庆，李华，刘慎权，李冬果.判定P3P问题正解数目的充要条件.计算机学报，1998，第21卷第12期，pp.1084-1095
[134]张彩霞. P3P问题多解的几个判定条件.北方工业大学学报，2007，第19卷第3期，pp.45-51
[135]张彩霞，胡占义. P3P问题的多解现象的概率研究.软件学报，2007，第18卷第9期，pp.2100-2104
[136]郝颖明，朱枫，欧锦军，吴清潇，周静，付双飞.基于点特征的位姿测量系统鲁棒性分析.计算机应用，2008，第28卷第7期，pp.1713-1716
[137]朱枫，郝颖明，欧锦军，吴清潇，周静，付双飞.横滚角对P3P位姿测量方法鲁棒性的影响分析.高技术通讯，2008，第18卷第10期，pp.1041-1046
[138]孙凤梅，王波.关于PnP问题多解的分布与解的稳定性的讨论.自动化学报，2010，第36卷第9期，pp.1213-1219