CGAL Accepted Projects page for the Google Summer of Code 2013As in 2010, 2011 and 2012, the CGAL Project is a mentoring organization of the Google Summer of Code. On this page we present some project ideas as well the information applicants have to provide us. GSoC applicants are welcome to propose other ideas and check if a mentor is interested in supervising it. Previous project ideas of the CGAL project for the Google Summer of Code can be found on the following pages: 2010, 2011, 2012, 2013. You will find below the list of projects that have been selected:Summary:
2D AABB TreeMentor: from Inria Sophia-Antipolis and from GeometryFactoryStudent: Nuwan Chamara Jayalath, Ensimag GrenobleINP - FRANCE Project DescriptionThe purpose of this project is to provide the same functionalities as the ones provided by the 3D AABB package but in 2D. Instead of duplicating of the code, the student will have to devise a solution that shares as much as possible source code from the 3D implementation, without breaking backward compatibility. Once completed, a testsuite carefully checking the correctness and performances of the implementation must be devised. The performance of such a core component being important, code profiling and optimization is a key aspect of the project. Time permitting, the student will propose a solution to the point-inside-polygon test using the 2D AABB-tree and will to participate to the ACM GISCUP 2013. Implement a model of the concept ArrangementDcel using the Combinatorial Map data structureMentors: from the Applied Computational Geometry Lab, Tel Aviv University and from LIRISStudent: Junfei Huang, The Ohio State University - USA Project DescriptionAn object of the class template More point generators in CGALMentor: from Federal University of Pernambuco.Student: Alexandru Tifrea, Politehnic University of Bucharest - Hungary Project DescriptionCGAL is a high quality generic C++ library on Computational Geometry. Amongst its features, CGAL offers random point generators in disks, squares, balls, cubes, hypercubes, to mention a few. Point generation is quite useful for several things, e.g., testing algorithms, obtaining initial values for optimization schemes, and more. In this work, we will concentrate efforts on implementing new point generators. Some of the useful point generators that are missing in CGAL include random points inside a triangle, simplex, triangulation, on a surface mesh, on a pure complex, ... Adding a Curve-Skeletonization Algorithm into CGALMentor: from Ecole Polytechnique Federale de Lausanne, and from GeometryFactory.Student: Xiang Gao, ETH Zurich - Switzerland Project DescriptionThis project consists in implementing a curve-skeletonization algorithm for a watertight surface mesh. The algorithm consists of two main phases. In the first step, a mass-spring system is solved, resulting in a progressive contraction of the model into a skeleton. In the final step, the watertight surface is converted into a network of curves by shortest-edge collapse. The basic algorithm can also be extended to guarantee that the curves are centered within the object. A visual demo of the algorithm is available here and the algorithm is explained in the following paper (a reference implementation is also available at this page):
Improved design and new features for the Mesh_2 packageMentor: from GeometryFactory, and from Inria Sophia-AntipolisStudent: Raul Gallegos Hidalgo, National University of San Agustin - Peru Project DescriptionThe Mesh_2 package implements a 2D isotropic triangle mesh generator. Currently, the domain to be meshed is defined by some constrained edges and a set of seed points. The constrained edges divide the plane into several connected components. The mesh domain is either the union of the bounded connected components including at least one seed, or the union of the bounded connected components that do no contain any seed. The project can be split into three phases:
or additionnal details, here is the documentation of the current Mesh_2 package and of the Mesh_3 package. Visibility Polygon ImplementationMentor: and Alexander Kröller from TU BraunschweigStudent (variant 1): Kan Huang, Stony Brook University - USA Student (variant 2): Francisc Bungiu, Bremen University - Germany Project DescriptionGiven a planar polygonal domain (which can be given as a polygon or as an arrangement) and a query point, compute the polygon of all points that can see the query point. Efficient algorithm implementations for this problem are still missing in CGAL. This project will therefore join forces with people at TU Braunschweig, with the purpose of developing a new package for visibility. Together we will work on implementing several different algorithm types (with vs. without preprocesing) for different problem inputs (simple polygons vs. polygons with holes vs. arrangements); including limited vision ranges would be a welcome extra. Adding point set consolidation algorithms into CGALMentor: from Inria Sophia-AntipolisStudent: Shihao Wu, South China University of Technology(SCUT) - China Project DescriptionPoint set processing is playing an important role in modern geometry processing due to the various input collected nowadays. This project consists in implementing into CGAL
Last modified on Monday, 01-Jul-2013 18:50:44 MEST.
contact information
|