低维矩阵约束的非刚体特征点分类方法

doi:10.3969/j.issn.1001-2400.2015.04.009

J4 ›› 2015, Vol. 42 ›› Issue (4): 53-56+69.doi: 10.3969/j.issn.1001-2400.2015.04.009

低维矩阵约束的非刚体特征点分类方法

刘侍刚1;辛晓萌1;彭亚丽1,2;裘国永1

(1. 陕西师范大学现代教学技术教育部重点实验室，陕西西安 710062；
2. 西安电子科技大学计算机学院，陕西西安 710071)

收稿日期:2014-03-12 出版日期:2015-08-20 发布日期:2015-10-12
通讯作者: 刘侍刚
作者简介:刘侍刚(1973-)，男，副教授，E-mail: shgliu@gmail.com．
基金资助:
国家自然科学基金资助项目(61402274，41171338，41271387)；中国博士后科学基金资助项目(20080430201)；中国博士后科学基金特别资助项目(200902594)；陕西师范大学中央高校基本科研业务费资助项目(GK201402040，GK201302029)；陕西师范大学学习科学交叉学科培育计划资助项目；陕西师范大学实验技术研究资助项目(SYJS201329，SYJS201314)；陕西师范大学校级重点教学改革研究资助项目(884407)

Low dimension matrix constraint method for classifying non-rigid features

LIU Shigang1;XIN Xiaomeng1;PENG Yali1,2;QIU Guoyong1

(1. Key Lab. of Modern Teaching Technology of Ministry of Education, Shaanxi Normal Univ., Xi'an 710062, China;
2. School of Computer Science and Technology, Xidian Univ., Xi'an 710071, China)

Received:2014-03-12 Online:2015-08-20 Published:2015-10-12
Contact: LIU Shigang

摘要/Abstract

摘要：

为了有效地对非刚体特征点进行分类，提出了一种低维子空间约束的非刚体特征点分类方法，该方法将所有非刚体特征点放入一个矩阵中，同时利用所有做刚体运动的特征点组成一个三维子空间的特性构造误差函数，误差最大的点可认为做非刚体运动，并逐一剔除差误，剩下的点即可认为做刚体运动．模拟实验和真实实验结果表明，该方法能够有效地实现非刚体特征点的分类．

关键词: 低维子空间, 特征点分类, 非刚体

Abstract:

To classify the features into the rigid and non-rigid ones, a low dimension matrix constraint method for non-rigid feature classification is proposed. All the non-rigid features are put into a matrix and an error function is introduced based on the fact that the matrix which consists of rigid features spans a 3D subspace. The feature which has a max error is regarded as the non-rigid one and moved from the matrix one by one. Experimental results with both simulated and real data show that the method can efficiently classify the features into two sets.

Key words: low dimension matrix, feature classified, non-rigid

刘侍刚;辛晓萌;彭亚丽;裘国永. 低维矩阵约束的非刚体特征点分类方法[J]. J4, 2015, 42(4): 53-56+69.

LIU Shigang;XIN Xiaomeng;PENG Yali;QIU Guoyong. Low dimension matrix constraint method for classifying non-rigid features[J]. J4, 2015, 42(4): 53-56+69.

参考文献

［1］ Peng Y, Liu F, Liu S. Active Contours Driven by Normalized Local Image Fitting Energy［J］. Concurrency and Computation: Practice and Experience, 2014, 26(5): 1200-1214.
［2］ Digne J, Steiner D, Alliez P, et al. Feature-preserving Surface Reconstruction and Simplification from Defect-laden Point Sets［J］. Journal of Mathematical Imaging and Vision, 2014, 48(2): 369-382.
［3］彭亚丽, 刘芳. 一种组织进化的遮挡点恢复算法［J］. 西安电子科技大学学报, 2013, 40(4): 137-141.
Peng Yali, Liu Fang. An Organizational Evolutionary Algorithm for Occlusion Recovery［J］. Journal of Xidian University, 2013, 40(4): 137-141.
［4］ Liu S, Peng Y. A Local Region-based Chan-Vese Model for Image Segmentation［J］. Pattern Recognition, 2012, 45(7): 2769-2779.
［5］ Sheehan M, Harrison A, Newman P. Automatic Self-calibration of a Full Field-of-view 3D n-laser Scanner［J］. Experimental Robotics, 2014, 79(1): 165-178.
［6］刘侍刚, 彭亚丽, 韩崇昭, 等. 基于秩1的射影重建方法［J］. 电子学报, 2009, 37(1): 225-228.
Liu Shigang, Peng Yali, Han Chongzhao, et al. Projective Reconstruction Based on Rank 1 Matrix［J］. Acta Electronica Sinica, 2009, 37(1): 225-228.
［7］彭亚丽, 刘侍刚, 刘芳. 基于秩1约束的三维重建方法［J］. 信号处理, 2010, 26(1): 28-31.
Peng Yali, Liu Shigang, Liu Fang. A 3D Reconstruction Method Based on Rank 1［J］. Signal Processing, 2010, 26(1): 28-31.
［8］ Zhou H, Li X, Sadka A. Nonrigid Structure-from-motion from 2-D Images Using Markov Chain Monte Carlo［J］. IEEE Transactions on Multimedia, 2012, 14(1): 168-177.
［9］ Lee M, Choi C. Real-time Facial Shape Recovery from a Single Image under General, Unknown Lighting by Rank Relaxation［J］. Computer Vision and Image Understanding, 2014, 120(3): 59-69.
［10］ Bregler C, Hertzmann A, Biermann H. Recovering Non-rigid 3D Shape from Image Streams［C］//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. Los Alamitos: IEEE, 2000: 690-696.
［11］ Alessio D. Adaptive Non-rigid Registration and Structure from Motion from Image Trajectories［J］. International Journal of Computer Vision, 2013, 103(2): 226-239.
［12］ Paladini M, Bue A, Dodig M, et al. Optimal Metric Projections for Deformable and Articulated Structure-from-motion［J］. International Journal of Computer Vision, 2012, 96(2): 252-276.
［13］ Lowe D. Distinctive Image Features from Scale-invariant Keypoints［J］. International Journal of Compute Vision, 2004, 60(2): 91-110.

[1]	余旺盛;侯志强;田孝华;查宇飞. 利用局部特征联合匹配的非刚体目标跟踪[J]. J4, 2014, 41(6): 160-166.
[2]	郑翔;同鸣;曹阳. 一种颅脑MR影像的LBM三维非刚体配准方法[J]. J4, 2013, 40(6): 125-131.
[3]	同鸣;曹阳;何梦玥;张建龙. LBM的医学影像非刚体配准新方法[J]. J4, 2013, 40(1): 6-13.

低维矩阵约束的非刚体特征点分类方法

Low dimension matrix constraint method for classifying non-rigid features

PDF (PC)

赞

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 3

Metrics

本文评价

推荐阅读 0