PCL Developers blog

Roadmap

  • Research various out-of-core spacial data structures and databases as well as investigate preexisting out-of-core techonologies.

    • Provide references for the development of the project
    • Design new file formats for storage of spacial data structure and databases
    • Implement point cloud processing tools for creating new file formats
  • Initial integration of Urban Robotics’s octree-based point cloud format in PCL

    • Get code compiling within PCL framework

    • Use Google Test instead of Boost’s Unit Test Framework

    • Refactor current code

      • Everything’s in a header
      • Remove old deprecated code
      • Cleanup cut/paste sections
      • Understand how everything works
      • Documentation
    • Replace underlying point data structures with PCL’s, ie PointType

    • Remove depenedency on cJSON library

  • Develop a barebones visualization UI with QVTK providing interactive frame rates of massive point clouds.

    • Learn (Q)VTK and the plausaility of using their framework out-of-the-box
    • May need to develop first class objects for direct rendering
    • Incorporate GLSL shaders.
    • Implement multiresolution rendering algorithms for reading in processed point clouds
  • Research aspects of parallel programming

  • ...