© The Institution of Engineering and Technology
A robust fitting algorithm for geometric computer vision problems under the L∞ -norm optimisation framework is presented. It is essentially based on the maximum feasible subsystem (MaxFS) but it overcomes the computational limitation of the MaxFS for large data by finding only a quasi-maximum feasible subset. Experimental results demonstrate that the algorithm removes outliers more effectively than the other parameter estimation methods recently developed when the outlier-to-inlier ratio in a data set is high.
References
-
-
1)
-
3. Yu, C., Seo, Y., Lee, S.W.: ‘Photometric stereo from maximum feasible lambertian reflections’. European Conf. on Computer Vision (ECCV), Heraklion, Crete, Greece, September 2010.
-
2)
-
4. Zheng, Y., Sugimoto, S., Okutomi, M.: ‘Deterministically maximizing feasible subsystem for robust model fitting with unit norm constraint’. IEEE Conf. on Computer Vision and Pattern Recognition (CVPR), Providence, RI, USA, June 2011.
-
3)
-
1. Lee, H., Seo, Y., Lee, S.W.: ‘Removing outliers by minimizing the sum of infeasibilities’, Image Vis. Comput., 2010, 28, (6), pp. 881–889 (doi: 10.1016/j.imavis.2009.11.004).
-
4)
-
2. Olsson, C., Eriksson, A., Hartley, R.: ‘Outlier removal using duality’. IEEE Conf. on Computer Vision and Pattern Recognition (CVPR), San Francisco, CA, USA, June 2010.
-
5)
-
5. Chinneck, J.: ‘Feasibility and infeasibility in optimization: Algorithms and computational methods’ (Springer, New York, 2007, 1st edn).
-
6)
-
7. Makhorin, A.: ‘GLPK (GNU linear programming kit)’, 2008. .
-
7)
-
6. Kahl, F., Hartley, R.: ‘Multiple-view geometry under the L∞-norm’, IEEE Trans. Pattern Anal. Mach. Intell., 2008, 30, (9), pp. 1603–1617 (doi: 10.1109/TPAMI.2007.70824).
http://iet.metastore.ingenta.com/content/journals/10.1049/el.2015.0842
Related content
content/journals/10.1049/el.2015.0842
pub_keyword,iet_inspecKeyword,pub_concept
6
6
Quasi-maximum feasible subsystem for geometric computer vision problems