基于全变分的运动分割模型及分裂 Bregman 算法

王诗言; 于慧敏

doi:10.16383/j.aas.2015.c140255

基于全变分的运动分割模型及分裂 Bregman 算法

doi: 10.16383/j.aas.2015.c140255

王诗言^1, ,,
于慧敏²

1.
重庆邮电大学重庆 400065;
2.
浙江大学信息与电子工程学系杭州 310027

基金项目:

国家重大科技专项(2014ZX03001027),国家重点基础研究发展计划(973计划)(2012CB316400),重庆邮电大学博士启动基金(A2014-09)资助

详细信息

作者简介:
于慧敏浙江大学信息与电子工程学系教授. 主要研究方向为图像处理和计算机视觉. E-mail: yhm2004@zju.edu.cn

通讯作者:
王诗言重庆邮电大学讲师. 2013 年获得浙江大学信息与通信工程专业博士学位. 主要研究方向为图像处理, 计算机视觉, 无线通信. 本文通信作者.E-mail: wangshiyan@cqupt.edu.cn

计量
- 文章访问数: 1911
- HTML全文浏览量: 133
- PDF下载量: 773
- 被引次数: 0
出版历程
- 收稿日期: 2014-04-17
- 修回日期: 2014-09-01
- 刊出日期: 2015-02-20

Motion Segmentation Model Based on Total Variation and Split Bregman Algorithm

WANG Shi-Yan^{1
, ,},
YU Hui-Min²

1.
Chongqing University of Posts and Telecommunications, Chongqing 400065;
2.
Department of Information and Electronics Engineering, Zhejiang University, Hangzhou 310027

Funds:

Supported by National Science and Technology Major Project (2014ZX03001027), National Basic Research Program of China (973 Program) (2012CB316400), and Ph.D. Foundation of Chongqing University of Posts and Telecommunications (A2014-09)

摘要

摘要: 提出了一种基于全变分的运动分割模型,可以适用于2D/3D视频.首先, 通过活动轮廓模型将分割与估计融合在同一能量函数中, 该模型能够同时进行分割曲面的演化和运动参数的估计. 其次,通过凸松弛方法将原始问题转化为等价的全变分模型, 克服了局部最小值问题.最后,采用分裂Bregman快速算法进行求解. 多组实验证明了本文方法对2D/3D视频的通用性和算法的高效性.
- 运动分割 /
- 运动估计 /
- 全变分 /
- 分裂Bregman算法
Abstract: A general motion segmentation model for 2D/3D videos based on total variation is presented in this paper. Firstly, a spatiotemporal energy functional based on active contour model is established to perform motion segmentation and estimation simultaneously. Secondly, a convex optimization technique is introduced to convert the original energy functional into a global convex functional based on total variation, which overcomes the local minimum problem. Finally, the Split Bregman algorithm is applied to solve the segmentation surface. Various experiments indicates the universality for both 2D and 3D videos, and demonstrates the feasibility and effectiveness of the proposed algorithm.
- Motion segmentation /
- motion estimation /
- total variation /
- split Bregman algorithm

HTML全文

参考文献(21)

[1]	Nikolov B, Kostov N, Yordanova S. Investigation of mixture of Gaussians method for background subtraction in traffic surveillance. International Journal of Reasoning-based Intelligent Systems, 2013, 5(3): 161-168
[2]	[2] Zhou X W, Yang C, Yu W C. Moving object detection by detecting contiguous outliers in the low-rank representation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2013, 35(3): 597-610
[3]	[3] Weinland D, Ronfard R, Boyer E. A survey of vision-based methods for action representation, segmentation and recognition. Computer Vision and Image Understanding, 2011, 115(2): 224-241
[4]	[4] Pinto A M, Correia M V, Paulo M A, Costa P G. Unsupervised flow-based motion analysis for an autonomous moving system. Image and Vision Computing, 2014, 32(6-7): 391-404
[5]	[5] Soheilian B, Paparoditis N, Vallet B. Detection and 3D reconstruction of traffic signs from multiple view color images. ISPRS Journal of Photogrammetry and Remote Sensing, 2013, 77: 1-20
[6]	[6] Wu Q B, Xiong J, Luo B, Wang Z N. A segmentation-based chroma intra prediction coding scheme for H.264/AVC. Circuits, Systems, and Signal Processing, 2014, 33(3): 939-957
[7]	[7] Whitaker T R. A level-set approach to 3D reconstruction from range data. The International Journal of Computer Vision, 1998, 29(3): 203-231
[8]	[8] Chan T F, Esedoglu S, Nikolova M. Algorithms for finding global minimizers of image segmentation and denoising models. SIAM Journal on Applied Mathematics, 2004, 66(5): 1632-1648
[9]	[9] Rudin L I, Osher S, Fatemi E. Nonlinear total variation based noise removal algorithms. Physica D: Nonlinear Phenomena, 1992, 60(1-4): 259-268
[10]	Mumford D, Shah J. Optimal approximations by piecewise smooth functions and associated variational problems. Communications on Pure and Applied Mathematics, 1989, 42(5): 577-685
[11]	Bresson X, Chan T F. Fast dual minimization of the vectorial total variation norm and applications to color image processing. Inverse Problems and Imaging, 2008, 2(4): 455-484
[12]	Zach C, Pock T, Bischof H. A duality based approach for realtime Tv-L1 optical flow. In: Proceedings of the 29th DAGM Conference on Pattern Recognition. Berlin, Heidelberg: IEEE, 2007. 214-223
[13]	Vese L A, Chan T F. A multiphase level set framework for image segmentation using the Mumford and Shah model. International Journal of Computer Vision, 2002, 50(3): 271-293
[14]	Mansouri A R, Mitiche A, EI-Feghali R. Spatio-temporal motion segmentation via level set partial differential equations. In: Proceedings of the 5th IEEE Southwest Symposium on Image Analysis and Interpretation, Santa Fe, USA: IEEE, 2002. 243-247
[15]	Wang S Y, Yu H M, Hu R. 3D video based segmentation and motion estimation with active surface evolution. Journal of Signal Processing Systems, 2013, 71(1): 21-34
[16]	Wang S Y, Yu H M. Primal-dual method for spatiotemporal tracking model with moving background. Journal of Zhejiang University (Engineering Science), 2013, 47(4): 630-637
[17]	Chan T F, Golub G H, Mulet P. A nonlinear primal dual method for total variation-based image restoration. SIAM Journal of Scientific Computing, 1999, 20(6): 1964-1977
[18]	Chambolle A. An algorithm for total variation minimization and applications. Journal of Mathematical Imaging and Vision, 2004, 20(1-2): 89-97
[19]	Bresson X, Esedoglu S, Vandergheynst P, Thiran J, Osher S. Fast global minimization of the active contour/snake models. Journal of Mathematical Imaging and Vision, 2007, 28(2):151-167
[20]	Bregman L M. The relaxation method of finding the common points of convex sets and its application to the solution of problems in convex programming. USSR Computational Mathematics and Mathematical Physics, 1967, 7(3): 200-217
[21]	Osher S, Burger M, Goldfarb D, Xu J J, Yin W T. An iterative regularization method for total variation-based image restoration. Multiscale Modeling and Simulation, 2005, 4(2): 460-489