Abstract
Consider a point-set coming from an object which was sampled using a digital sensor (depth range, camera, etc). We are interested in finding a graph that would represent that point-set according to some properties. Such a representation would allow us to match two objects (graphs) by exploiting topological properties instead of solely relying on geometrical properties. The Delaunay triangulation is a common out off-the-shelf strategy to triangulate a point-set and it is used by many researchers as the standard way to create the so called data-graph and despite its positive properties, there are also some drawbacks. We are interested in generating a graph with the following properties: the graph is (i) as unique as possible, (ii) and as discriminative as possible regarding the degree distribution. We pose a combinatorial optimization problem (Min-Weight Max-Entropy Problem) to build such a graph by minimizing the total weight cost of the edges and at the same time maximizing the entropy of the degree distribution. Our optimization approach is based on Dynamic Programming (DP) and yields a polynomial time algorithm.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
Conrad, K.: Probability distributions and maximum entropy (2005), http://www.math.uconn.edu/~kconrad/blurbs/analysis/entropypost.pdf
Cross, A.D.J., Hancock, E.R.: Graph matching with a dual-step em algorithm. IEEE Transactions on Pattern Analysis and Machine Intelligence 20(11), 1236–1253 (1998), ISSN 0162-8828
de Sousa, S., Kropatsch, W.G.: Graph-based point drift: Graph centrality on the registration of point-sets. Pattern Recognition 48(2), 368–379 (2015), ISSN 0031-3203
Delaunay, B.N.: Sur la sphère vide. Bulletin of Academy of Sciences of the USSR, 793–800 (1934)
Dickinson, P.J., Bunke, H., Dadej, A., Kraetzl, M.: Matching graphs with unique node labels. Pattern Analysis and Applications 7(3), 243–254 (2004), ISSN 1433-7541
Erdös, T., Gallai, P.: Gráfok előírt fokszámú pontokkal. Matematikai Lapok 11, 264–274 (1960)
Gabriel, K.R., Sokal, R.R.: A new statistical approach to geographic variation analysis. Systematic Biology 18(3), 259–278 (1969)
Hakimi, S.: On realizability of a set of integers as degrees of the vertices of a linear graph. i. Journal of the Society for Industrial and Applied Mathematics 10(3), 496–506 (1962)
Havel, V.: Poznámka o existenci konečných graf•367u. Časopis Pro Pěstování Matematiky 080(4), 477–480 (1955)
Kocay, W., Kreher, D.L.: Graphs, Algorithms, and Optimization. Discrete Mathematics and Its Applications. Taylor & Francis (2004) ISBN 9780203489055
Lian, W., Zhang, L.: Rotation invariant non-rigid shape matching in cluttered scenes. In: Daniilidis, K., Maragos, P., Paragios, N. (eds.) ECCV 2010, Part V. LNCS, vol. 6315, pp. 506–518. Springer, Heidelberg (2010)
Luo, B., Hancock, E.R.: Iterative procrustes alignment with the {EM} algorithm. Image and Vision Computing 20(5-6), 377–396 (2002), ISSN 0262-8856
Ohrhallinger, S., Mudur, S.: An efficient algorithm for determining an aesthetic shape connecting unorganized 2d points. Computer Graphics Forum 32(8), 72–88 (2013), ISSN 1467-8659
Reeb, G.: Sur les points singuliers d’une forme de pfaff complétement intégrable ou d’une fonction numérique. C. R. Acad. Sci. Paris 222, 847–849 (1946)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2015 Springer International Publishing Switzerland
About this paper
Cite this paper
de Sousa, S., Kropatsch, W.G. (2015). Data Graph Formulation as the Minimum-Weight Maximum-Entropy Problem. In: Liu, CL., Luo, B., Kropatsch, W., Cheng, J. (eds) Graph-Based Representations in Pattern Recognition. GbRPR 2015. Lecture Notes in Computer Science(), vol 9069. Springer, Cham. https://doi.org/10.1007/978-3-319-18224-7_2
Download citation
DOI: https://doi.org/10.1007/978-3-319-18224-7_2
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-18223-0
Online ISBN: 978-3-319-18224-7
eBook Packages: Computer ScienceComputer Science (R0)