3D Ear Shape Feature Optimal Matching Using Bipartite Graph
In this paper, we present an optimized matching algorithm based on bipartite graph for 3D ear shape key points. Comparing with the graph matching algorithm of key points, our algorithm avoid the 2D Delaunay triangulation on 3D key points, then has less accuracy error; and our complexity is lower because our matching algorithm is basing on the bipartite graph. And then we optimal the bipartite graph matching work by weighting the edge between the key points. Experiments show that, our optimal matching on bipartite graph of ear key points can get a higher matching accuracy and a better matching efficiency.
KeywordsEar matching Keypoints Shape feature Bipartite optimal matching
This work is supported in part by National Natural Science Foundation of China with projects No. 61170143 and No. 60873110.
- 1.Burge M., Burger W (2000) Ear biometrics in computer vision. In: Proceedings 15th international conference on pattern recognition, vol 2, pp 822–826. doi: 10.1109/ICPR.2000.906202
- 2.Torresani L, Kolmogorov V, Rother C (2008) Feature correspondence via graph matching: models and global optimization. In Proceedings 10th European conference computer vision, pp 596–609. doi: 10.1007/978-3-540-88688-4_44
- 4.Leordeanu M, Hebert M (2009) Unsupervised learning for graph matching. In: IEEE conference on computer vision and pattern recognition, pp 864–871. doi: 10.1109/CVPRW.2009.5206533
- 10.SasiKanth (2010) Bi-dimensional empirical mode decomposition. MATLAB central file exchange selectGoogle Scholar