Outlier removal using duality
作者： Carl Olsson , Anders Eriksson , Richard Hartley , None
DOI: 10.1109/CVPR.2010.5539800
关键词:
摘要: In this paper we consider the problem of outlier removal for large scale multiview reconstruction problems. An efficient and very popular method task is RANSAC. However, as RANSAC only works on a subset images, mismatches in longer point tracks may go undetected. To deal with would like to have, post processing step RANSAC, that entire (or larger) part sequence. two algorithms doing this. The first one related by Sim & Hartley where quasiconvex solved repeatedly error residuals largest removed. Instead solving each show it enough solve single LP or SOCP which yields significant speedup. Using duality same theoretical result holds our method. second algorithm faster version first, L 1 -optimization. While well practice, there no guarantee success. We these methods are through duality, evaluate number data sets promising results.1
uq.edu.au参考文章(19)
Yongduek Seo, Hyunjung Lee, Sang Wook Lee, Outlier Removal by Convex Optimization for L-Infinity Approaches pacific-rim symposium on image and video technology. pp. 203- 214 ,(2009) , 10.1007/978-3-540-92957-4_18
Jos F. Sturm, Using SeDuMi 1.02, a MATLAB toolbox for optimization over symmetric cones Optimization Methods & Software. ,vol. 11, pp. 625- 653 ,(1999) , 10.1080/10556789908805766
Qifa Ke, Takeo Kanade, Quasiconvex Optimization for Robust Geometric Reconstruction IEEE Transactions on Pattern Analysis and Machine Intelligence. ,vol. 29, pp. 1834- 1847 ,(2007) , 10.1109/TPAMI.2007.1083
Fredrik Kahl, Richard Hartley, None, Multiple-View Geometry Under the {$L_\infty$}-Norm IEEE Transactions on Pattern Analysis and Machine Intelligence. ,vol. 30, pp. 1603- 1617 ,(2008) , 10.1109/TPAMI.2007.70824
David L. Donoho, For most large underdetermined systems of equations, the minimal 1-norm near-solution approximates the sparsest near-solution Communications on Pure and Applied Mathematics. ,vol. 59, pp. 907- 934 ,(2006) , 10.1002/CPA.20131
Sameer Agarwal, Noah Snavely, Steven M. Seitz, Fast algorithms for L ∞ problems in multiview geometry computer vision and pattern recognition. pp. 1- 8 ,(2008) , 10.1109/CVPR.2008.4587713
E.J. Candes, T. Tao, Decoding by linear programming IEEE Transactions on Information Theory. ,vol. 51, pp. 4203- 4215 ,(2005) , 10.1109/TIT.2005.858979
R.I. Hartley, In defence of the 8-point algorithm international conference on computer vision. pp. 1064- 1070 ,(1995) , 10.1109/ICCV.1995.466816
Arnak Dalalyan, Renaud Keriven, L_1-Penalized Robust Estimation for a Class of Inverse Problems Arising in Multiview Geometry neural information processing systems. ,vol. 22, pp. 441- 449 ,(2009)
J.A. Tropp, Just relax: convex programming methods for identifying sparse signals in noise IEEE Transactions on Information Theory. ,vol. 52, pp. 1030- 1051 ,(2006) , 10.1109/TIT.2005.864420