Transactions on Computational Science XX

The main objective in content-based image retrieval is to find images similar to a query image in an image collection. Matching using descriptors computed from regions centered at local invariant interest points (keypoints) have become popular because of their robustness to changes in viewpoint and occlusion. However, local descriptor matching can produce many false matches. To improve the retrieval results, geometric verification is usually performed as a post-pocessing step. RANSAC can robustly fit a model to data in presence of outliers and has been widely used for the geometric verification stage. But obtaining a good hypothesis may require many trial runs, particularly when the proportion of inliers in the data is low. We introduce a novel geometric verification scheme called DT-RANSAC based on topological information in the Delaunay Triangulation of putatively matched keypoints to construct a refined set of matches, that is presented to the RANSAC algorithm to fit a homography. Experiments reveal that DT-RANSAC is able to converge to correct hypothesis in very few trial runs and the retrieval results are consistently better than geometric verification based on plain RANSAC.

[1]  Deok-Soo Kim,et al.  Quasi-triangulation and interworld data structure in three dimensions , 2006, Comput. Aided Des..

[2]  Richard M. Karp,et al.  Two special cases of the assignment problem , 1975, Discrete Mathematics.

[3]  Edith Cohen,et al.  Recognizing Properties of Periodic Graphs , 1990, Applied Geometry And Discrete Mathematics.

[4]  Daniel Baum,et al.  Voronoi-Based Extraction and Visualization of Molecular Paths , 2011, IEEE Transactions on Visualization and Computer Graphics.

[5]  Deok-Soo Kim,et al.  Euclidean Voronoi diagram of 3D balls and its computation via tracing edges , 2005, Comput. Aided Des..

[6]  Richard M. Karp,et al.  The Organization of Computations for Uniform Recurrence Equations , 1967, JACM.

[7]  Deok-Soo Kim,et al.  Querying simplexes in quasi-triangulation , 2012, Comput. Aided Des..

[8]  P. Gonçalves,et al.  Free energy of solvation from molecular dynamics simulation applying Voronoi-Delaunay triangulation to the cavity creation. , 2005, The Journal of chemical physics.

[9]  Darrah Chavey,et al.  Tilings by regular polygons—II: A catalog of tilings , 1989 .

[10]  Atsuyuki Okabe,et al.  Spatial Tessellations: Concepts and Applications of Voronoi Diagrams , 1992, Wiley Series in Probability and Mathematical Statistics.

[11]  A. Bondi.  van der Waals Volumes and Radii , 1964 .

[12]  Hiroshi Imai,et al.  Proximity and Motion Planning on l_1-Embeddable Tilings , 2011, 2011 Eighth International Symposium on Voronoi Diagrams in Science and Engineering.

[13]  Dominique Attali,et al.  Modeling noise for a better simplification of skeletons , 1996, Proceedings of 3rd IEEE International Conference on Image Processing.

[14]  Deok-Soo Kim,et al.  Molecular surfaces on proteins via beta shapes , 2007, Comput. Aided Des..

[15]  Th. Hahn,et al.  Nomenclature and generation of three-periodic nets: the vector method , 1984 .

[16]  Robert L. Scot Drysdale,et al.  Voronoi diagrams based on convex distance functions , 1985, SCG '85.

[17]  Michel Deza,et al.  Geometry of cuts and metrics , 2009, Algorithms and combinatorics.

[18]  Michel Deza,et al.  Clin D'oeil on L1-embeddable Planar Graphs , 1997, Discret. Appl. Math..

[19]  Norie Fu,et al.  A Strongly Polynomial Time Algorithm for the Shortest Path Problem on Coherent Planar Periodic Graphs , 2012, ISAAC.

[20]  Deok-Soo Kim,et al.  Protein-Ligand Docking Based on Beta-Shape , 2010, Trans. Comput. Sci..

[21]  Gabriella Sanniti di Baja,et al.  Pruning Discrete and Semiocontinuous Skeletons , 1995, ICIAP.

[22]  Grégoire Malandain,et al.  Euclidean skeletons , 1998, Image Vis. Comput..

[23]  Frédéric Chazal,et al.  The "lambda-medial axis" , 2005, Graph. Model..

[24]  Changhee Lee,et al.  BetaDock: Shape-Priority Docking Method Based on Beta-Complex , 2011, Journal of biomolecular structure & dynamics.

[25]  Deok-Soo Kim,et al.  Anomalies in quasi-triangulations and beta-complexes of spherical atoms in molecules , 2013, Comput. Aided Des..

[26]  János Pach,et al.  Reconfigurations in Graphs and Grids , 2006, SIAM J. Discret. Math..

[27]  Jong Bhak,et al.  Sphericity of a protein via the beta-complex. , 2010, Journal of molecular graphics & modelling.

[28]  Michael T. Goodrich,et al.  Voronoi diagrams for convex polygon-offset distance functions , 2001, Discret. Comput. Geom..

[29]  Michel Deza,et al.  Scale-isometric polytopal graphs in hypercubes and cubic lattices - polytopes in hypercubes and Zn , 2004 .

[30]  Longin Jan Latecki,et al.  Discrete Skeleton Evolution , 2007, EMMCVPR.

[31]  J. Eon,et al.  Infinite geodesic paths and fibers, new topological invariants in periodic graphs. , 2007, Acta crystallographica. Section A, Foundations of crystallography.

[32]  Farid Karimipour,et al.  A Stable Voronoi-based Algorithm for Medial Axis Extraction through Labeling Sample Points , 2012, 2012 Ninth International Symposium on Voronoi Diagrams in Science and Engineering.

[33]  Derick Wood,et al.  Voronoi Diagrams Based on General Metrics in the Plane , 1988, STACS.

[34]  Kenneth Steiglitz,et al.  Optimization of one-bit full adders embedded in regular structures , 1986, IEEE Trans. Acoust. Speech Signal Process..

[35]  Hiroshi Imai,et al.  Voronoi Diagrams on Periodic Graphs , 2010, 2010 International Symposium on Voronoi Diagrams in Science and Engineering.

[36]  Mariette Yvinec,et al.  Algorithmic geometry , 1998 .

[37]  Farid Karimipour,et al.  Voronoi-Based Curve Reconstruction: Issues and Solutions , 2012, ICCSA.

[38]  Deok-Soo Kim,et al.  Beta‐decomposition for the volume and area of the union of three‐dimensional balls and their offsets , 2012, J. Comput. Chem..

[39]  Kunwoo Lee,et al.  Principles of CAD/CAM/CAE Systems , 1999 .

[40]  Marina L. Gavrilova,et al.  Implementation of the Voronoi-Delaunay Method for Analysis of Intermolecular Voids , 2004, ICCSA.

[41]  Deok-Soo Kim,et al.  Quasi-worlds and quasi-operators on quasi-triangulations , 2010, Comput. Aided Des..

[42]  Masayuki Abe,et al.  Drift-compensated data acquisition performed at room temperature with frequency modulation atomic force microscopy , 2007 .

[43]  Deok-Soo Kim,et al.  Three-dimensional beta-shapes and beta-complexes via quasi-triangulation , 2010, Comput. Aided Des..

[44]  Martti Mäntylä,et al.  Introduction to Solid Modeling , 1988 .

[45]  Mark Pauly,et al.  The scale axis transform , 2009, SCG '09.

[46]  Deok-Soo Kim,et al.  Triangulation of molecular surfaces , 2009, Comput. Aided Des..

[47]  Michael O'Keeffe,et al.  Crystal nets as graphs: Terminology and definitions , 2005 .